Accreditations
Tuition fee EU nationals (2026/2027)
Tuition fee non-EU nationals (2026/2027)
Check here the Daytime detailed study plan
Check here the Evening detailed study plan
The undergraduate degree in Data Science is based on the convergence of different scientific areas - Mathematics, Statistics and Informatics – and its programme structure is structured around projects which foster both practical and theoretical thinking, with a view towards granting the student an informed, critical, and autonomous understanding of data in the face of the various dimensions of the Knowledge Society and the Digital Revolution.
The Bachelor's is well-situated for helping students to comprehend and explore the areas of this knowledge-base. These actions support the student's progressive acquisition of independence and the capacity to respond to problems of increasing complexity.
With the synthesis, which occurs in the last two semesters, the coherence of the training program is consolidated around responsible practice and the exceptional professional skills required in order to respond to the challenges of modern society.
Programme Structure for 2026/2027
Data in Science, Bussiness and Society
After the course the student should be able to achieve the Learning Outcomes (LO):
OA1: Account for different definitions of data, different data types and different research approaches that generate it.
OA2: Identify the knowledge claims underlying different interpretations of data.
OA3: Explain the difference between quantitative and qualitative approaches to data generation.
OA4: Examine the implications of data collection for research, business and society.
OA5: Discuss different debates about the implications of data for people in organizations and society.
CP1: What data are and how to think with data.
CP2: Types of problems addressed in Data Science and specificities in the domains of Science, Management, and Society.
CP3: Different traditions and research methodologies and definitions of knowledge acquisition.
CP4: Translating real challenges into technical concepts and using scientifically oriented language.
CP5: The ethical dimension of data use strategies.
CP6: Presentation of practical cases.
This course uses only assessment throughout the semester and does not include exams.
Assessment components:
a) Mini-tests (30%): 6 mini-tests (5% each, the vast majority to be taken at home)
b) Project (30%): group assignment
c) Final test (40%): Written test to be taken during the 1st season, 2nd season or special season (Art. 14, RGACC)
Passing requirement: Final test >= 8 points (out of 20 points)
The final grade for the Project will depend on the code, the reports, and the student's performance in presenting their work.
Cathy O'Neil, Rachel Schutt, Doing Data Science: Straight Talk from the Frontline, 2014, ISBN: 9781449358655,
Borgman, C. L., Big data, little data, no data: scholarship in the networked world, 2015, ISBN: 9780262529914,
Rob Kitchin, The data revolution: Big data, open data, data infrastructures and their consequences, 2014, https://doi.org/10.4135/9781473909472,
Davenport, T., Harris, J., and Morison, R., Analytics at work: smarter decisions, better results. Harvard Business Review Press, USA., 2010, ISBN: 9781422177693,
Turban, E., Sharda, R., Delen, D., Decision Support and Business Intelligence Systems (9th Eds), 2010, ISBN: 978-0136107293,
Davenport, T., Big Data at Work: Dispelling the Myths, Uncovering the Opportunities, 2014, ISBN: 978-1422168165,
Calculus Topics II
At the end of the course, each student should be able to:
LG1. Compute partial derivatives and gradients.
LG2. Determine linear approximations of functions of several variables.
LG3. Determine and classify critical points of functions of several variables.
LG4. Apply the previous concepts in the context of regression problems.
LG5. Compute double integrals.
LG6. Apply integral calculus to the evaluation of volume, mass and probability.
LG7. Interpret geometrically all the previous concepts.
LG8. Implement in MATLAB some of the computacional methods studied in class.
PC1. Differential calculus
1.1. Limits and continuity
1.2. Partial derivatives.
1.3. Tangent plane and differentiability.
1.4. The chain rule
1.5. Computation and classification of critical points.
1.6. Gradient descent.
1.7. Linear regression.
PC2. Integral calculus.
2.1. Double integral.
2.2. Double integrals in polar coordinates.
2.3. Application of integral calculus to the evaluation of volume, mass and probability.
Students must obtain an overall grade of at least 10 (out of 20) in one of the assessment modes:
- Assessment throughout the semester: 2 Written Tests (40% each) + participation in class (20%).
- A final Exam (100%) in either the 1st or 2nd examination period.
Stewart, J. "Cálculo - Volume 2", Tradução da 8ª edição norte-americana (4ª edição brasileira), Cenage Learning, 2017.
Exploratory Data Analysis
Learning goals (LG) to be developed in articulation with the general objectives:
LG1. Prepare data for analysis.
LG2. Use and interpret a set of statistical tools in the field of descriptive.
LG3. Use Excel, R and Jamovi in data preparation, analysis and representation applications.
LG4. Adapt the visual representation models to different objectives, according to good visualization practices.
LG5. Interpreting and writing the results of a descriptive data analysis.
Syllabus contents (SC) articulated with the learning objectives.
SC1. Organization, preparation and transformation of data
SC2. Exploratory data analysis
Missing values
Coding and imputation
Exploratory charts
Random variables
Empirical distribution function
Normal Distribution
SC3. Descriptive data analysis
Descriptive measures
Single and bivariate analysis
Association measures
SC4. Visual representation
Introduction to the principles of visual representation
Visual representation structures
Assessment throughout the semester:
- Individual exercise with R (10%)
- 3 mini-tests to be taken during lessons, with the two highest marks being considered for the weighting in the overall assessment (10% + 10%)
- Group work, with presentation and discussion (30%); minimum grade 7.5
- Written test (40%); min. grade 7.5
A minimum attendance of 70% of class hours is required for assessment throughout the semester.
Assessment by exam:
- Exam/individual practical work (35%); minimum grade 7.5
- Written exam (65%); minimum grade 7.5
Brown, D.S. (2022). Statistics and Data Visualization Using R. The Art and Practice of Data Analysis. Sage Publication, Inc.
Cairo, A. (2013). The Functional Art: An introduction to information graphics and visualization (Voices That Matter).
New Riders.
Carvalho, A. (2017). Métodos quantitativos com Excel, Lisboa, Lidel edições técnicas.
Chang, W. (2024) R Graphics Cookbook. 2nd ed. O’Reilly. (Disponível em: https://r-graphics.org/)
Reis, E. (1998). Estatística Descritiva, Lisboa, Sílabo,7ª ed.
Rocha, M. & Ferreira, P.G. (2017) Análise e Exploração de Dados com R. Lisboa, FCA
Alexandrino da Silva, A. (2006). Gráficos e mapas-representação de informação estatística. Lisboa, Lidel edições técnicas.
Barroso, M., Sampaio, E. & Ramos, M. (2003). Exercícios de Estatística Descritiva para as Ciências Sociais, Lisboa, Sílabo.
Carvalho, A. (2017). Gráficos com Excel - 95 Exercícios, Lisboa, FCA.
Dias Curto, J.J., & Gameiro, F. (2016). Excel para Economia e Gestão. Lisboa, Ed. Sílabo.
Field, A., Miles, J., & Field, Z. (2012). Discovering statistics using R. London, SAGE Publications Ltd.
Hoaglin, D.C., Mosteller, F & Tukey, J. W. (1992). Análise Exploratória de Dados. Técnicas Robustas, Ed. Salamandra, Lisboa.
Murteira, B. J. F. (1990). Análise Exploratória de Dados. Estatística Descritiva, McGraw Hill.
Wickham, H. (2015). ggplot2: Elegant Graphics for Data Analysis, Springer. (Disponível em: https://ggplot2-book.org/)
Data Structures and Algorithms
At the end of this course, students should be able:
LO1: Identify, rewrite, and examine common forms of data organization and its associated algorithms (with and without dynamic memory management, with iterative or recursive algorithms);
LO2: Identify the most appropriate and efficient data structure for a problem;
LO3: Know how to evaluate and compare the order of performance and efficiency of a given algorithm and/or data structure for the common operations of inserting, removing, and accessing;
LO4: Understand the pros and cons of recursive, and iterative algorithms, as well as dynamic programming.
LO5: Understand different search and sorting algorithms appropriate for computational solutions.
CP1: Data Structures and Algorithms: what are these and why are they important? Abstract Data Types
CP2: Linear data structures: stacks, queues, double-ended queues, lists, and linked lists.
CP3: Introduction to algorithm complexity analysis.
CP4: Search algorithms: linear, and binary search.
CP5: Recursion, iteration, backtracking, and dynamic programming.
CP6: Iterative sorting algorithms: Bubblesort, Selectionsort, and Insertionsort.
CP7. Recursive sorting algorithms: Mergesort, and Quicksort.
CP8: Nonlinear data structures: tree, binary search trees, AVL trees, and graphs.
CP9: Simple algorithms for nonlinear data structures.
It has been approved by the ISTA Pedagogical Committee that the EDA assessment is done only through assessments throughout the semester (see Observations 1). Therefore, it is not possible to assess EDA with an exam worth 100%.
Minimum attendance criteria: 75% of the 12 weeks x 3 hours of class.
Assessment elements and their respective weightings in the final grade:
- Individual written test -> 60%, minimum grade of 7.5 points, plaaned to be held in the after-work class of the last week of classes;
- Individual assignment 1, with oral exam -> 15%;
- Individual assignment 2, or in groups of 2 (to be decided), with individual oral exam -> 25%, minimum grade of 7.5 points.
The final grade is obtained with the weighted average: Final_grade = 60% x Test_grade + 15% x Assignment_1_grade + 25% x Assignment_2_grade.
The classifications of the 3 assessment elements are used in the formula without rounding to the nearest whole number. To pass the course, the final grade must be greater than or equal to 10 points.
Students must be familiar with the Student Disciplinary Regulations (see Observations 2) and fulfill their general duties, as described in point 2 of Article 2 of those regulations (see Observations 2).
Students with special status conferred by the Serviços de Gestão do Ensino (contact these Services, as professors are not competent to grant this status) may have access to the special period (see Observations 3). In the Special Period, the assessment elements and their respective weightings in the final grade are:
- individual written exam -> 60%, minimum grade of 7.5 points;
- two individual assignments, with an individual oral exam -> 15% + 25%, minimum grade of 7.5 in each of the assignments.
The final grade in the special period is obtained in the same way as in the regular period (assessment throughout the semester).
- J. Wengrow, A Common-Sense Guide to Data Structures and Algorithms, Second Edition. The Pragmatic Bookshelf, 2020.
- M. Goodrich, R. Tamassia, and M. Goldwasser, Data Structures & Algorithms in Python. Wiley, 2013.
- B. Miller and D. Ranum, Problem Solving with Algorithms and Data Structures using Python, Second Edition, Release 3.0. 2013.
- T. Cormen, C. Leiserson, R. Rivest, and C. Stein, Introduction to Algorithms, Fourth Edition. MIT Press, 2022.
- Referências adicionais a indicar durante as aulas.
Calculus Topics I
At the end of this course the student should be able to:
LO1: Calculate limits of sequences
LO2. Compute derivatives and interpret the corresponding result.
LO3. Determine linear and higher order approximations.
LO4. Explicitly compute the antiderivative of some elementary functions.
LO5. Use the fundamental theorem of calculus to differentiate functions defined by integrals.
LO6. Use integrals to compute areas, lengths, probabilities, etc.
LO7. Integrate some notable ODEs.
LO8. Apply some simple numerical methods to compute approximate values and use graphical computational tools.
1. Sequences
1.1. Some concepts
1.2. Convergence
1.3. Some limits and useful results
2. Differential calculus in R
2.1. A brief review
2.2. Continuity and limits
2.3. Differentiability and Taylor’s formula; applications
2.4. Numerical methods
2.4.1. Fixed point method
2.4.2. Bisection method
2.4.3. Newton-Raphson method
2.4.4. Numerical differentiation
3. Integral calculus in R
3.1. Anti-derivatives
3.2. Integrals
3.3. Fundamental theorem of calculus
3.4. Numerical integration
3.4.1. Mid-point method
3.4.2. Trapezoidal rule
4. Ordinary differential equations.
4.1. Separable variables
4.2. First order linear equations
4.3. Numerical methods
4.3.1. Euler’s method
4.3.2. Runge-Kutta method (RK4)
A student must obtain an overall grade of at least 10 (out of 20) in one of the following assessment modes:
- Assessment during the semester: a mid-term test (37.5%) + final test (37.5%) + two group works, one about numerical calculus and the other about graphical representation (25%, 12.5% each).
- Exam assessment: in any of the exam seasons (100%).
The exam consists of two parts: analytical (75%) and numerical (25%). Students who have successfully completed the teamwork may skip this numerical component.
The minimum grade is 8. Students with a grade over 16 should be submitted to an oral examination.
[1] Ferreira, J.C. (2011). “Introdução à Análise Matemática”, Fundação Calouste Gulbenkian.
[2] Strang, G. (1991). “Calculus”, Wellesley-Cambridge.
[3] Caputo, H.P. (1973). “Iniciação ao Estudo das Equações Diferenciais”, Livros Técnicos e Científicos Editora, S.A.
[4] Suleman, A., Rocha, J., Alho, A., Apontamentos de aula. (disponível no Moodle)
[5] Suleman, A., Notas sobre cálculo numérico (disponível no Moodle).
[6] Santos, M.I.R., Matemática computacional (IST).
Optimization for Data Science
By successfully completing this Curricular Unit, students will be able to:
LO1. Develop formulations in linear programming, integer linear programming, and non-linear programming for efficiently solve complex problems in real contexts.
LO2. Interpret outputs obtained by general software for solving problems formulated in linear programming, integer linear programming, and non-linear programming.
LO3. Do the economic interpretation and produce recommendations based on the results obtained for problems formulated in linear programming, integer linear programming, and non-linear programming.
Programmatic Contents (PC):
PC1: Linear Programming
1.1 General form of a Linear Programming model
1.2 Formulating problems in Linear Programming
1.3 Graphical resolution
1.4 Resolution using general software (Excel Solver or other)
1.5 Interpreting results and sensitivity analysis
PC2: Integer Linear Programming
2.1 Formulating problems in Integer Linear Programming
2.2 Formulating problems with binary variables
2.3 Resolution using general software (Excel Solver or other)
2.4 Interpreting results
2.5 Branch-and-Bound algorithm
PC3: Non-linear Programming
3.1 Formulating problems in Nonlinear Programming
3.2 Resolution using general software (Excel Solver or other)
3.3 Interpreting results
Assessment throughout the semester or assessment by exam:
1. Assessment throughout the semester:
a) 5 individual short tests: i) weighting 10%; ii) completed in class throughout the lecturing period; iii) grade for this component: arithmetic mean of the 4 best grades obtained on the 5 short tests;
b) Group project: i) weighting 30%; ii) Groups of 4 students; iii) with oral discussion;
c) Individual final written test: i) weighting 60%; ii) minimum classification of 8.5;
d) Attendance of at least 2/3 of the classes taught;
e) Approval: minimum weighted average of 9.5.
2. Evaluation by Exam (1st and 2nd Season):
a) Individual written exam: i) weight of 100%; ii) approval: minimum classification of 9.5.
In both assessment methods, students may be required to take an individual exam, either oral or written, aimed at validating the grade obtained.
Scale: 0-20 points.
* Ragsdale, C.T. (2017). Spreadsheet Modeling and Decision Analysis: A Practical Introduction to Business Analytics. 8th Ed. Cengage Learning.
* Evans, J. (2021). Business Analytics. 3rd Ed. Global Edition. Pearson.
* Hillier, F.S and Lieberman, G.J. (2015). Introduction to Operations Research, 10th Ed., McGraw-Hill.
* Ragsdale, C.T. (2001). Spreadsheet Modeling & Decision Analysis: A Practical Introduction to Management Science. 3rd Ed., South-Western College Publishing.
* Wolsey, L.A. (1998). Integer Programming. Wiley.
Probabilities and Sampling
At the end of the curricular unit, students should be able to simulate probabilistic scenarios (LO1), calculate conditioned probabilities (directly or via Bayes' theorem), and verify the independence of events (LO2). They must know the
classical statistical sampling methods and their applicability conditions (LO3) and know how to calculate estimates and their precision measures (LO4). They should be able to identify contexts of complementarity between survey sampling and big data (LO5).
CP1. Probability theory: definitions, axioms, conditional probability, total probability and Bayes' formula.
CP2. Univariate random variables: probability and density function, distribution function, parameters.
CP3. Survey methodology: data collection by questionnaire; sampling plan.
CP4. Random vs. Non-random sampling: types of non-random sampling.
CP5. Random sampling: simple, systematic, with unequal probabilities, stratified.
CP6. Combining surveys with big data.
Assessment regime: throughout the semester or by exam.
Throughout the semester:
- Mandatory attendance of at least 75% of the classes (18 classes).
- Midterm test (50%) + final test (50%).
- Both tests have a minimum grade of 7.5 values and consist of two components: theoretical-interpretive (70%) and applied assessment with R (30%).
By exam:
- Exam (100%) consists of two components: theoretical-interpretive (70%) and applied assessment with R (30%).
Whenever there are doubts regarding the authorship or level of the knowledge demonstrated in an assessment component, the student may be required to take an individual oral or written test to validate the corresponding grade. An oral exam may also be required if the grade is 18 or higher; if the student does not attend the oral exam, the final grade will be 16 values.
The use of cell phones, smartwatches, or any other electronic devices is not permitted during written exams. Unauthorized use of such devices will be penalized in accordance with the Student Disciplinary Regulations.
Groves, R., Fowler, F., Couper, M., Lepowski, J., Singer, E. & Tourangeau, R. (2009) Survey Methodology, 2nd edition, John Wiley and Sons.
Levy, P., Lemeshow, S. (1991). Sampling of Populations: methods and applications. Wiley Interscience.
Reis, E., Andrade, M., Calapez, T. & Melo, P., Estatística Aplicada, volume 1. 7ª edição. Lisboa. Edições Sílabo, 2021, ISBN 978-989-561-186-7.
Reis, E., Andrade, M., Calapez, T. & Melo, P., Estatística Aplicada volume 2, 6ª edição, Lisboa. Edições Sílabo, 2019, ISBN 978-972-618-986-2.
Speegle, D., & Clair, B. (2021). Probability, Statistics, and Data: A Fresh Approach Using R. Chapman and Hall/CRC. Free access at https://mathstat.slu.edu/~speegled/_book/
Salganik, M. (2018). Bit by Bit- Social Research in the Digital Age. New Jersey: Princeton University Press Verzani, J. (2014). Using R for Introductory Statistics, 2nd Edition, Chapman & Hall/CRC, eBook ISBN 9781315373089. https://cran.r-project.org/doc/contrib/Verzani-SimpleR.pdf
Programming
After obtaining approval in the course, students should be able to:
OA1. Develop functions/procedures that implement simple algorithms.
OA2. Develop code that manipulates arrays and objects.
OA3. Develop simple object classes, considering the notion of encapsulation.
OA4. Write and understand Python code.
CP1. Variables and control structures
CP2. Functions and parameters
CP3. Invocation and recursion
CP4. Procedures and input/output
CP5. Objects and references
CP6. Object classes
CP7. Composite objects
CP8. Composite object classes
CP9. Arrays
CP10. Matrices
This course is done only by assessment throughout the semester, not considering the modality of assessment by exam.
Evaluation components:
a) HWs (15%): 6 online mini-tests, to do at home;
b) TEST1 (20%): Closed-book written test during the intermediate evaluation week;
c) PROJECT (25%): Individual project;
d) TEST2 (40%): Closed-book written test during the 1st season, 2nd season or special season (Art. 14 RGACC).
Approval requirement: TEST2 >= 8 points.
The final grade for the PROJECT is determined for each student through an oral exam or practical test, which uses the project submitted by the student as a basis to assess the acquired knowledge.
Other relevant information:
- Attendance is not an essential requirement for approval
- Questions asked in the written tests may involve aspects related to the project;
- Whenever there are doubts regarding the authorship or the mastery of the knowledge demonstrated in an assessment component, the student may be summoned to an individual oral or written exam intended to validate the respective grade;
- It is not possible to pass only by taking the final exam;
- in case of failure in the 1st season, the student can take TEST2 in the 2nd season, keeping the grade of the other components;
- When the grade improvement occurs in a school year different from the one in which the work was done, the grade of the components PROJECT, HWs and TEST1 is replaced by a practical test, to be performed on a computer before or after the written exam. Students under these conditions who wish to improve their grades should contact the course coordinator in advance, at least 2 days before the 1st season.
João P. Martins, Programação em Python: Introdução à programação com múltiplos paradigmas, 2013, IST Press, https://istpress.tecnico.ulisboa.pt/produto/programacao-em-python-introducao-a-programacao-utilizando-multiplos-paradigmas/
Big Data Processing
At the end of the course, students should be able to:
OA1: understand and know the main platforms for processing large amounts of information
OA2: understand and know how to apply distributed programming/computing models
OA3: understand the stages associated with a machine learning project for large amounts of information
OA4: know how to apply dimensionality reduction techniques
OA5: apply supervised or unsupervised learning techniques to large-scale problems
CP1: Computing platforms for big data
CP2: Machine learning pipeline for big data
CP3: Dimensionality reduction
CP4: Supervised/unsupervised learning for large scale
CP5: Case studies: PageRank and Recommendation Systems
This course includes the following assessment methods: (1) assessment throughout the semester; (2) assessment by exam.
(1) Assessment throughout the semester
The final grade is made up of:
- Individual written test (70%), with a minimum mark of 8.0;
- Group work (30%).
The group work has a mid-term submission, which will count for 30%, and a submission at the end of the semester, which will count for 70%. Those who do not submit the mid-term portion will automatically be assessed by exam.
The work will include an oral presentation/discussion, and the final grade will be individual.
(2) assessment by exam
The final grade will be based on a single written exam, including special season.
- Practical Data Science with Hadoop and Spark: Designing and Building Effective Analytics at Scale, Ofer Mendelevitch, Casey Stella and Douglas Eadline, Addison-wesley, 2016.
- Advanced Analytics with Spark: Patterns for Learning from Data at Scale, Sandy Ryza et al., O'Reilly Media, 2017.
- Learning Spark: Lightning-Fast Big Data Analysis, Holden Karau, A. Konwinski, P. Wendell and M. Zaharia, O'Reilly Media, 2015.
- Big Data: Algorithms, Analytics, and Applications, Kuan-Ching Li et al., Chapman and Hall/CRC, 2015.
- Mining of Massive Datasets, A. Rajaraman, J. Ullman, 2011, Cambridge University Press.
- The elements of statistical learning, Trevor Hastie, Robert Tibshirani, and Jerome Friedman. Springer, 2001
- All of Statistics: A concise course in Statistical Inference, L.Wasserman, Springer, 2003.
Fundamentals of Database Management
O1: Develop abstraction mechanisms;
O2: Develop Information Modeling abilities;
O3: Develop the ability to extract data from a database in an efficient way.
P1 - Database Design
P2 Relations and primary keys
P1.2.2 Foreign Keys and Integrity Rules
P1.2.3 Optimizationsand Indexes
P1.2.5 Transctions and Concurrency
P2 S.Q.L
P2. 1 Simpl Querys;
P2.2 Agregate Functions;
P2.3 SubQuerys;
P2.4 Triggers and Stored Procedures;
Assessment is done through exam, sseason 1, season 2 and special season.
Bibliography-Ramos, P, Desenhar Bases de Dados com UML, Conceitos e Exercícios Resolvidos, Editora Sílabo, 2ª Edição, 2007
-Perreira, J. Tecnologia de Base de Dados" FCA Editora de Informática, 1998
-Damas, L. SQL - Structured Query Language " FCA Editora de Informática, 2005 (II)
http://plsql-tutorial.com/.
-Date, C.J. "An introduction to Database Systems" Addison-Wesley Publishing Company, sexta edição, 1995 (I.2, I.3, I.4, II);
-Booch, G., Rumbaugh, J., Jacobson, I "The Unified Modeling Language User Guide" Addison-Wesley Publishing Company, 1999 (I.1);
-Nunes, O´Neill, Fundamentos de UML, FCA, 2002
Regression Models
LO1. Understand the correlation between variables and simple and multiple linear regression models.
LO2. Know the main parameter estimation methods (OLS and ML).
LO3. Know how to perform regression model assumption analysis, hypothesis testing, and diagnosis.
LO4. Understand the limitations of linear regression using quantile regression and regularization methods.
LO5. Application of the concepts studied; training/test and prediction sets, information and value extraction for real-world data with R software.
S1. Introduction to Regression Models
S1.1. Correlation
S1.2. Simple linear regression
S1.3. Multiple linear regression
S2. Estimation and inference
S2.1 OLS estimation
S2.2 ML estimation
S2.3 Statistical inference
S3. Model diagnosis and validation
S3.1. Assumptions
S3.2 Diagnostic Tests
S3.3. Practical Examples
S4. Extensions of Linear Regression
S4.1 Regularisation Methods (Ridge and Lasso)
S4.2. Quantile Regression
S4.3. Practical Examples
S5. Applications using R for real-world data and scenarios
Assessment throughout the semester includes the following:
a) An individual mid-term test, accounting for 40% of the mark.
b) An individual test, accounting for 60% of the mark, with a minimum mark of 8.5.
Exam assessment: an individual exam covering the entire syllabus, with a minimum mark of 10 (mark rounded to the nearest whole number). The individual test and the exam will be held without the use of handouts, books or other materials; the use of graphing calculators or mobile phones is not permitted; students may only consult the form and tables made available on Moodle for this purpose.
In both cases, students may be called upon to sit an individual oral or written assessment to validate their mark.
Cameron, A.C., Trivedi, P.K. (2008) Microeconometrics: Methods and Applications. Cambridge University Press
Bruce P., Bruce A., and Gedeck P., (2020), Practical Statistics for Data Scientists: 50+ Essential Concepts Using R and Python, 2nd Edition,
O' Reilly Media, Inc. - Robert I. Kabacoff, (2022), R in Action: Data analysis and graphics with R, Third Edition, Manning Publications Co. -
Tom Alby, (2024), Data Science in Practice, CRC Press.
Robert I. Kabacoff, (2022), R in Action: Data analysis and graphics with R, Third Edition, Manning Publications Co. Wooldridge, J.M. (2019), "Introductory Econometrics: A Modern Approach", 7th Ed., Cengage Learning.
Eric Goh Ming Hui, (2019), Learn R for Applied Statistics, Apress.
Daniel J. Denis, (2020), Univariate, Bivariate, and Multivariate Statistics Using R: Quantitative Tools for Data Analysis and Data Science, JohnWiley & Sons, Inc. Ficheiros (slides e scripts) da UC a disponibilizar no Moodle.
Heuristic Optimization
At the end of the Curricular Unit, the student is expected to be able to:
LO1- Discuss challenges faced in real, large scale optimization problems
LO2 - Explain and discuss the available methodologies for addressing hard optimization problems
LO3 - Formulate and design effective solution methods for addressing optimization problems
LO4- Employ the use of advanced tools to solve optimization problems
Programmatic contents (PC):
PC1. MULTIOBJECTIVE PROGRAMMING
1.1. Basic concepts
1.2. Methodologies
1.2.1. Non-Preemptive Goal Programming
1.2.2. Preemptive Goal Programming
PC2. METAHEURISTICS
2.1. Concepts and terminology
2.2. Single point algorithms
2.2.1. Tabu Search
2.2.2. Simulated Annealing
2.3. Genetic Algorithms
1st SEASON ASSESSMENT
In the 1st Season, the Course can be assessed using the Assessment throughout the Semester or by completing an Individual Project.
--> ASSESSMENT THROUGHOUT THE SEMESTER
- Intermediate Test (35%): minimum mark of 7,5;
- Group Project (65%): written report and code (40%) + oral presentation (10%) + individual test (15%).
Conditions associated with assessment throughout the semester:
(i) Maximum number of students who can make up a working group: 5;
(ii) The student must participate in all moments of assessment throughout the semester;
(iii) Group Project: minimum mark of 7,5.
--> INDIVIDUAL EVALUATION
The individual assessment consists of a Test and an Individual Project:
- Test (35%), to be taken on the day of the exam (first season): minimum mark of 7,5
- Individual Project (65%; minimum mark of 7,5):
- Written report and code (40%)
- Oral discussion (25%).
The student must take part in all components of the individual assessment.
2nd SEASON ASSESSMENT
In the 2nd Season, the evaluation is based on a Test and an Individual Project.
--> INDIVIDUAL EVALUATION
- Test (35%), to be taken on the day of the exam (2nd season): minimum mark of 7,5
- Individual Project (65%; minimum mark of 7,5):
- Written report and code (40%)
- Oral discussion (25%).
The student must take part in all components of the individual assessment.
Passing the course: A final mark of 10 or above.
In both forms of assessment, students may be required to take an additional individual evaluation, either oral or written, in order to confirm their mark.
Scale 0-20
- Chopard, B.; Tomassini, M. (2018). An Introduction to Metaheuristics for Optimization, Springer.
- Gutierrez, A. M; Ramirez-Mendoza, R. A.; Flores, E. M.; Ponce-Cruz, P; Espinoza, A.A. O.; Silva, D. C. B. (Eds.) (2020). A Practical Approach to Metaheuristics using LabVIEW and MATLAB (R). Taylor & Francis Ltd.
- Ke-Lin Du; M. N. S. Swamy (2016). Search and Optimization by Metaheuristics: Techniques and Algorithms Inspired by Nature. Birkhäuser.
- Ragsdale, C.T. (2022). Spreadsheet Modeling and Decision Analysis: A Practical Introduction to Business Analytics, 9th Ed. Cemgage Learning.
- Taha, Handy A. (2023). Operations Research: An Introduction, 11th Ed. Pearson. (não tem Goal Programming e tem heurísticas)
- Taha, Handy A. (2017). Operations Research: An Introduction, 10th Ed. Pearson. (tem Goal Programming e tem heurísticas)
- Burke, E. K.; Kendall, G. (Eds.) (2014). Search Methodologies: Introductory Tutorials in Optimization and Decision Support, 2nd edition, Springer.
- Ehrgott, M. (2005). Multicriteria Optimization, 2nd edition, Springer.
- Lobato, F. S.; Valder, S. Jr. (2017). Multi-Objective Optimization Problems: Concepts and Self-Adaptive Parameters with Mathematical and Engineering Applications. Springer Cham.
- Open Access documents such as instructor notes, book chapters, research articles, and tutorials that will be provided via Fenix.
- Siarry, P. (Ed.) (2016). Metaheuristics, Springer.
Security, Ethics and Privacy
LG1. Recognize the main security issues in software-based systems, their causes, and consequences.
LG2. Identify and describe the security services necessary to implement a specific information protection policy based on risk analysis.
LG3. Learn the principles and regulatory frameworks in the domains of personal data protection and privacy, with special focus on the General Data Protection Regulation of 2016.
LG4. Ethically and critically reflect on the implications of technologies and data processing on individuals and society, addressing the resulting challenges in the fields of information security, data protection, and privacy.
CP1. Information Security:
Fundamentals of security – data security;
Vulnerabilities and threats in security;
IRM – Information Risk Management;
Cryptography – symmetric, asymmetric, hash functions, MAC (Message Authentication Code), digital signatures;
Public Key Infrastructures (PKI).
CP2. Privacy and Personal Data Protection:
The GDPR and Law 58/2019;
Anonymization and pseudonymization techniques.
CP3. Ethics:
Normative ethical theories;
Regulatory dilemmas and gaps in ICT;
Case studies in applied ethics;
The regulation of Artificial Intelligence
The course unit provides for two assessment modes: assessment throughout the semester and assessment by examination.
Assessment throughout the semester:
Assessment throughout the semester includes:
* Individual tests: throughout the semester, students will take short individual tests, lasting approximately 15–25 minutes each, administered in person and covering material from previous classes. The number of tests may vary depending on the pace of the classes, with these tests accounting collectively for 66% of the final grade.
* A group assignment, carried out within the Ethics module, accounting for 34% of the final grade.
A minimum grade of 7 out of 20 is required in the group assignment. The assignment is assessed on the basis of the content, clarity and consistency of the analysis.
Group assignments with a provisional grade of 17 out of 20 or above are subject to a group oral discussion. The discussion forms part of the assessment of the assignment, and the grade may be adjusted for the group as a whole or individually.
Assessment by examination:
Assessment by examination consists of an individual written examination, accounting for 100% of the final grade, to be taken during the official assessment period and covering all content taught in the course unit. Where necessary, it may include a supplementary oral examination.
In the first examination period, students may choose between assessment throughout the semester and assessment by examination. Students who choose assessment throughout the semester and do not pass may only take the examination in the second examination period.
Use of artificial intelligence:
In the group assignment, the use of artificial intelligence tools is governed by Iscte’s Guide for the Responsible Use of Artificial Intelligence in Academic Work.
Whenever AI tools are used, their use must be declared in accordance with the Guide and accompanied by the applicable institutional form. Students remain fully responsible for the content submitted.
Andress, J. (2014). The Basics of Information Security: Understanding the Fundamentals of InfoSec in Theory and Practice. Syngress.
Kim, D., Solomon, M. (2016). Fundamentals of Information Systems Security. Jones & Bartlett Learning.
Cannon, J.C. Privacy in Technology: Standards and Practices for Engineers and Security and IT Professionals. Portsmouth: AN IAPP Publication, 2014.
Breaux, Travis. Introduction to IT Privacy: A Handbook for Technologists. Portsmouth: An IAPP Publication, 2014.
Whitman, M., & Mattord, H. (2013). Management of information security. Nelson Education.
Katz, J., & Lindell, Y. (2014). Introduction to modern cryptography. CRC press.
Ethics, Technology, and Engineering: An Introduction (2023), Ivovan de Poel, Lamber Royakkers, Wiley-Blackwell, 2nd edition
European Union Agency for Fundamental Rights, The Handbook on European data protection law, 2019, http://fra.europa.eu/sites/default/files/fra_uploads/fra-coe-edps-2018-handbook-data-protection_en.pdf
Floridi, L. (2019). The Logic of Information: A Theory of Philosophy as Conceptual Design. Oxford University Press.
Canto Moniz, G. (2023). Manual de introdução à proteção de dados pessoais. Coimbra: Almedina.
Whitman, M., Mattord, H. (2017). Principles of Information Security. Course Technology.
Bowman, Courtney. The Architecture of Privacy: On Engineering Technologies that Can Deliver Trustworthy Safeguards. O?Reilly Media, 2015.
Anderson, R. J. (2010). Security engineering: a guide to building dependable distributed systems. John Wiley & Sons.
Zúquete, A. (2018). Segurança em redes informáticas. FCA-Editora de Informática.
Regulamentos e orientações da Comissão Europeia relativos à Proteção de Dados, https://www.edpb.europa.eu/our-work-tools/general-guidance/guidelines-recommendations-best-practices_pt
Bynum, Terrell Ward, and Simon Rogerson, (2004), Computer Ethics and Professional Responsibility: Introductory Text and Readings. Oxford: Blackwell, 2004.
Grupo do Artigo 29, Parecer 05/2014 sobre técnicas de anonimização do grupo de trabalho de proteção de dados do artigo 29.º, de 10 de Abril de 2014, 2014, -, https://ec.europa.eu/justice/article-29/documentation/opinion-recommendation/files/2014/wp216_pt.pdf
Enisa, Orientações da Enisa sobre técnicas de pseudonimização e boas práticas, 2019, -, https://www.enisa.europa.eu/publications/pseudonymisation-techniques-and-best-practices
UE, Proposta do regulamento do parlamento europeu e do conselho que estabelece regras harmonizadas em matéria de inteligência artificial (regulamento inteligência artificial) e altera determinados atos legislativos da União, 2023, -, https://eur-lex.europa.eu/legal-content/PT/TXT/?uri=CELEX%3A52021PC0206
Outros textos a indicar e distribuídos pelo docente ao longo do semestre.
Unsupervised Learning Methods
LO1: Characterize the main unsupervised data methods
LO2: Use R for unsupervised data analytics
LO3: Evaluate, validate and interpret the results
PC1: Introduction to unsupervised learning methods
PC2: Data reduction techniques (dimensionality)
- Principal components analysis (PCA)
- Data reduction techniques using R
PC3: Clustering techniques
- Hierarchical methods
- Partitioning methods
- Self-organizing maps
- Probabilistic methods
- Quality & Validity of clustering methods
- Clustering techniques using R
PC4: Case studies
Students may choose either Evaluation during the semester or Final Exam.
EVALUATION DURING THE SEMESTER:
- group project with discussion and minimum grade of 8 (50%)
- individual test with minimum grade of 8 (50%)
Approval requires a minimum attendance of 80% of classes, participation in all assessment components, and minimum grade of 10.
EXAM:
The Final Exam is a written exam. Students must achieve a minimum grade of 10 to pass.
In both forms of assessment, students may be called upon to sit an individual examination, either oral or written, to validate their grade.
Nwanganga, F., M. Chapple (2020), Practical Machine Learning in R, 1st Edition, Wiley. Bouveyron, C., G. Celeux, T. B. Murphy, A. E. Raftery (2019), Model-Based Clustering and Classification for Data Science: With Applications in R, 1st Edition, Cambridge University Press. James, G., Witten, D., Hastie, T., Tibshirani, R. (2013), An Introduction to Statistical Learning: with applications in R, New York: Springer. Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E. (2014), Multivariate Data Analysis, 7th Edition, Essex, UK: Pearson Education.
-
Network Analysis
By successfully completing this Curricular Unit, students will be able to:
LO1. Classify networks using correlation and clustering coefficients, distances, centrality measures and heterogeneity measures. Evaluate the network robustness.
LO2. Obtain the co-occurrence network associated with a network representing relations. Analyze weighted networks.
LO3. Select random network generation models and characterize random networks.
LO4. Detect communities and evaluate the methods applied to detect communities.
Programmatic Contents (PC):
PC1: Basic Concepts
Elements of a network, subnetworks, density and degree. Bipartite networks.
PC2: Small Worlds
Degree correlation. Paths and distances. Connectivity. Six Degrees of Separation. Clustering coefficients.
PC3: Hubs and Weight Heterogeneity
Centrality measures. Heterogeneity based on degree. Robustness, core decomposition and weight heterogeneity.
PC4: Random Networks
Generation and characterization of random networks.. Watts-Strogatz’s model, Configuration Model, and preferential models.
PC5: Communities
Concepts and related problems. Methods for community detection: bridge removal, modularity optimization, label propagation. Evaluation methods.
Assessment throughout the semester or assessment by exam:
1. Assessment throughout the semester:
a) 5 individual short tests: i) weighting 10%; ii) completed in class throughout the lecturing period; iii) grade for this component: arithmetic mean of the 4 best grades obtained on the 5 short tests;
b) Group project: i) weighting 30%; ii) Groups of 4 students; iii) with oral discussion;
c) Individual final written test: i) weighting 60%; ii) minimum classification of 8.5;
d) Attendance of at least 2/3 of the classes taught;
e) Approval: minimum weighted average of 9.5.
2. Evaluation by Exam (1st and 2nd Season):
a) Individual written exam: i) weight of 100%; ii) approval: minimum classification of 9.5;
In both assessment methods, students may be required to take an individual exam, either oral or written, aimed at validating the grade obtained.
Scale: 0-20 points.
* Menczer, F., Fortunato, S. and Davis, C. A. (2020). A First Course in Network Science, 1st Edition, Cambridge University Press, Cambridge.
* Barabási, A.-L. (2016). Network Science, 1st Edition, Cambridge University Press, Cambridge.
* Newman, M. (2018). Networks, 2nd Edition. Oxford University Press, Oxford.
* Ognyanova, K. (2018). Introduction to R and Network Analysis. SC&I Methods Workshop, Rutgers University.
Big Data Storage
LO1 - Implement data storage solutions in distributed and fault-tolerant support;
LO2 - Manipulation and extraction of large amounts of information from unstructured databases;
LO3 - Develop soft skills, namely problem solving, teamwork and collaboration, and critical observation.
CP1 - Introduction to Non-Relational Databases;
CP2 - Redundancy to manage fault tolerance;
CP3 - Data distribution to manage large volumes of information;
CP4 - Introduction to MongoDB;.
CP5 - Database Design in MongoDB;
CP6 - Manipulation of JSON structures;
CP7 - Data extraction in MongoDB.
Assessment throughout the semester is done through a written test (minimum grade: 7.5) which takes place on the same date as the 1st season exam and which is worth 60% of the grade and a group work, 40% of the grade (minimum grade: 7.5), to be delivered in the last week of classes.
Alternatively, there is assessment by exam - 100% of the grade (1st season, 2nd season and special season).
NoSQL Database: New Era of Databases for Big data Analytics - Classification, Characteristics and Comparison, A B M Moniruzzaman, Syed Akhter Hossain, 2013 (https://arxiv.org/abs/1307.0191)
MongoDb Homepage
Computational Statistics
Learning goals (LG) to be developed :
LG1: Consolidate the use of R software in the RStudio environment
LG2: Know how to calculate probabilities in various contexts, including through simulation
LG3: Be familiar with the most common probabilistic behavior models
LG4: Know how to fit probabilistic models
LG5: Understand the principles of statistical inference
LG6: Know how to choose the most appropriate inferential method for each situation
Syllabus contents (SC):
SC1- Probability theory: definitions, axioms, conditional probability, total probability theorem and Bayes' formula
SC2- Univariate random variables: mass and density functions, distribution function, and parameters. Working with usual random variables. Simulation of RV with a specified distribution.
SC3-Bi and multivariate RVs. Joint probability and distribution functions. correlation and covariation. Independence between RVs. Sample joint distribution.
SC4- Sampling distributions: limit central theorem, theoretical sampling distributions.
SC5- Parameters estimation: point estimation, estimators' properties, maximum likelihood estimators, interval estimation
SC6- Hypothesis testing: types of errors and corresponding probabilities. Test for one and two means. Chi-square of independence. Meaning and computation of p-values.
Students may choose either Assessment Throughout the Semester or Final Exam.
Assessment Throughout the Semester
1. Small assessment components throughout the semester**: 15% of the final grade. These may include quizzes, homework exercises, contributions to the Glossary or Forum on Moodle, etc. About 10 contributions are expected, with the best 8 being counted.
2. Midterm test, weight: 30%, no minimum grade required.
3. Final test, weight: 55%, minimum grade of 8 out of 20 required.
OR
Final Exam, 100%
A student who, during the regular exam period, has not achieved a passing grade but has scores for #1 (ongoing assessment components throughout the semester) and #2 (midterm test), may use the resit exam period (or special exam period, if applicable) as a replacement for the final test. Their final grade in these periods will be calculated according to the rules of assessment throughout the semester. The student must state this intention in writing on the exam they take.
This option is not available for grade improvement purposes.
Speegle, D., & Clair, B. (2021). Probability, Statistics, and Data: A Fresh Approach Using R (1st ed.). Chapman and Hall/CRC. Free access at https://mathstat.slu.edu/~speegled/_book/
Reis, E., Andrade, M., Calapez, T. & Melo, P., Estatística Aplicada, volume 1. 6ª edição. Lisboa. Edições Sílabo., 2015, ISBN 978-972-618-819-3.
Reis, E., Andrade, M., Calapez, T. & Melo, P., Estatística Aplicada volume 2, 6ª edição, Lisboa. Edições Sílabo., 2016, ISBN 978-972-618-986-2.
Verzani, J., Using R for Introductory Statistics, 2nd Edition, Chapman & Hall/CRC, 2014, eBook ISBN 9781315373089, https://cran.r-project.org/doc/contrib/Verzani-SimpleR.pdf
Reis, E., Andrade, M., Calapez, T. & Melo, P., Exercícios de Estatística Aplicada volume 1. 2ª edição, Lisboa. Edições Sílabo., 2012, ISBN 978-972-618-688-5
Reis, E., Andrade, M., Calapez, T. & Melo, P., Exercícios de Estatística Aplicada volume 2. 2ª edição, Lisboa. Edições Sílabo., 2014, ISBN 978-972-618-747-9
Supervised Learning Methods
LG1: Understanding supervised learning methods: scopes of application and procedures
LG2: Use of R software to perform data analysis
LG2: Evaluate and interpret the data analysis results
PC1: Overview of Supervised Learning
Typologies
Learning data
Objective functions
Models' assessment and selection
Notes on statistical inference
PC2: Regression Methods
K-Nearest Neighbor
Regression Trees (using CART algorithm)
PC3: Classification Methods
Naive Bayes
K-Nearest Neighbor
Logistic Regression
Classification Trees (using CART algorithm)
The Course can be assessed using the Assessment throughout Semester or Assessment by Exam.
ASSESSMENT THROUGHOUT SEMESTER:
- group quiz online (40%) with a minimum grade of 9
- individual test (60%) with a minimum grade of 9
Approval requires a minimum grade of 10.
ASSESSMENT BY EXAM:
1st part - individual test (60%)
2nd part -individual practical data analysis test, online, with the R software used in classes (40%)
Students have to achieve a minimum grade of 9 in each part of the exam and a combined minimum grade of 10.
Scale 0-20
Gareth, J., Daniela, W., Trevor, H., & Robert, T. (2013). An introduction to statistical learning: with applications in R. Springer.
Hastie, T., Tibshirani, R., Friedman, J. H., & Friedman, J. H. (2009). The elements of statistical learning: data mining, inference, and prediction (Vol. 2, pp. 1-758). New York: Springer.
Lantz, B. (2023). Machine Learning with R: Learn techniques for building and improving machine learning models, from data preparation to model tuning, evaluation, and working with big data. 4th Edition. Packt Publishing.
Molnar, C. (2020). Interpretable machine learning. Lulu. com.
Field, A., Miles, J., & Field, Z. (2012). Discovering statistics using R (2012). Great Britain: Sage Publications, Ltd, 958.
Larose, D., Larose, C. (2015). Data Mining and Predictive Analytics. John Wiley & Sons.
Introduction to Deep Learning
LO1: Explain the fundamental principles of deep learning and its evolution within the field of artificial intelligence.
LO2: Formulate deep learning problems by identifying data, models, parameters, loss functions, evaluation metrics, and generalization criteria.
LO3: Describe and analyze dense neural networks, including activation functions, depth, function composition, representational capacity, expressiveness, and practical limitations.
LO4: Train neural networks using optimization methods, understanding the role of backpropagation, weight initialization, and training stability.
LO5: Evaluate and improve the performance of deep learning models, identifying limitations, risks of overfitting and underfitting, and trade-offs between performance, interpretability, and computational cost.
LO6: Understand modern architectures, including convolutional neural networks, residual networks, Transformers, pre-trained models, and generative models such as VAEs and diffusion models.
P1 — Deep Learning Foundations: machine learning context; learning paradigms; data, models, parameters, loss functions, and generalization.
P2 — Neural Networks: shallow and deep networks; activation functions; depth, representation, and expressiveness.
P3 — Network Training: loss functions; gradient-based optimization; backpropagation; and weight initialization.
P4 — Evaluation and Regularization: training, validation, and testing; metrics; overfitting, underfitting, weight decay, dropout, data augmentation, and early stopping.
P5 — Convolutional Architectures: convolution, filters, pooling, CNNs, residual networks, transfer learning, and encoder–decoder models.
P6 — Transformers: tokenization, embeddings, attention, Transformers, fine-tuning, inference, and large language models.
P7 — Generative Models: autoencoders, latent spaces, VAEs, reconstruction, sampling, and diffusion models.
The course may be completed through one of two assessment methods:
1 – Continuous Assessment
Continuous assessment consists of:
- (80%) Two tests: one midterm test, held during the semester on a date to be announced, and a second test held on the first examination date;
- (20%) Online quizzes based on the practical assignments, completed through the e-learning platform. Attendance at the corresponding practical session is mandatory for the assessment to be considered.
The final grade is calculated as a weighted average of these two components according to the percentages above, subject to the following conditions:
- Students must obtain a minimum grade of 7.5/20 in both tests to pass the test component;
- Absence from three or more practical sessions results in automatic failure of the practical assessment component;
- For the purpose of the final grade calculation, practical assignments corresponding to missed sessions are awarded a grade of zero.
Failure in either component of the continuous assessment requires the student to take the final examination assessment.
2 – Final Examination Assessment
The final examination assessment consists of a single exam taken during any of the available examination periods and accounts for 100% of the final grade.
At any time, and regardless of the assessment method, students may be required to undertake an oral examination to confirm the awarded grade, verify the authorship of submitted work, and/or demonstrate the competencies assessed in the course.
- Prince, S. J. D. (2023). Understanding Deep Learning. MIT Press. (acesso aberto: https://udlbook.github.io/udlbook/)
- Zhang, A., Lipton, Z. C., Li, M., & Smola, A. J. (2023). Dive into Deep Learning. (acesso aberto: https://d2l.ai)
- Raschka, S., Liu, Y., & Mirjalili, V. (2022). Machine Learning with PyTorch and Scikit-Learn. Packt.
- Tunstall, L., von Werra, L., & Wolf, T. (2022). Natural Language Processing with Transformers. O'Reilly.
- Goodfellow, I., Bengio, Y., & Courville, A. (2016). *Deep Learning*. MIT Press.
- Documentação oficial do PyTorch e tutoriais da Hugging Face.
Stocastic Modelling
On completion of this course, students should:
LG1. Understand the principles and methods of stochastic simulation;
LG2. Be able to develop efficient algorithms for generating pseudorandom numbers;
LG3. Be able to apply the Monte Carlo method;
LG4. Understand and be able to apply different Monte Carlo via Markov Chains methods;
LG5. Be able to implement resampling techniques;
LG6. Be able to simulate a real system through discrete event simulation;
LG7. Be able to analyze and evaluate simulation results;
LG8. Be able to implement efficient stochastic simulation algorithms in R.
S1. Introduction to Simulation in Data Science
S2. Generation of Pseudo-Random Numbers
- Linear Congruential Method; Inverse Transformation Method; Acceptance/Rejection Method; Other Transformations; Mixtures
S3. Monte Carlo Methods in Statistical Inference
S4. Markov Chain Monte Carlo (MCMC) Methods
S5. Resampling Methods
- Bootstrap; Cross-Validation
S6. Discrete Event Simulation
- Assessment throughout the semester
Assessment throughtout the semester requires attendance at a minimum of two-thirds of the scheduled classes and comprises the following components:
• Delivery of two classes by students, working in groups, on topics assigned by the lecturer – 40%
• Class participation and contribution – 10%
• Individual written test – 50%
To pass the course, students must obtain a minimum mark of 8.5/20 in the written test and an overall final mark of at least 10/20.
- Assessment by Final Examination
Students opting for assessment by final examination will be assessed on the basis of:
• Written examination – 50%
• Individual essay – 50%
The essay should demonstrate critical analysis, a sound understanding of the course content, and the appropriate application of the methods and techniques covered in the course.
- Supplementary Oral Examination
The Course Unit Coordinator may require a supplementary oral examination in relation to any assessment component specified in this Course Unit Outline whenever this is deemed necessary to confirm the acquisition of the knowledge and competences being assessed or to clarify discrepancies between different assessment components.
The supplementary oral examination does not constitute a separate assessment component, carries no independent weighting in the final course mark, and serves exclusively to validate the mark awarded for the relevant assessment component. The mark previously awarded to that component shall be regarded as provisional until the supplementary oral examination has taken place and may be confirmed or revised in accordance with the outcome of the validation process and the applicable assessment criteria.
Students selected for a supplementary oral examination will be notified at least 24 hours in advance, with the date and time of the examination specified. Failure to attend without an acceptable justification will prevent validation of the relevant assessment component, and the consequences applicable to non-completion or withdrawal from that assessment component shall apply.
Templ, M. (2016). Simulation for Data Science with R. Packt Publishing Ltd:Birmingham, Uk.
Rizzo, M. L. (2008). Statistical Computing with R. Chapman & Hall/CRC.
Robert, C. P. and Casella, G. (2010). Introducing Monte Carlo Methods with R. Springer-Verlag.
Wickham, H. and Grolemund , G. (2017). R for Data Science. O'Reilly Media Inc.
Applied Final Project in Data Science
At the end of the course, each student should be able to:
LO1. Define the objectives and formulate the CD tasks that allow the client to extract the desired knowledge.
LO2. Define the data variables and metadata that lead to the required knowledge.
LO3. Plan the different phases of project development.
LO4. Process the data with the most appropriate Data Science tools to achieve the proposed objectives.
LO5. Produce data visualisations and documents suitable for correctly communicating the results obtained.
LO6. Solve problems inherent in using real data from an ethics-by-design perspective.
The programme contents (CP) are as follows:
CP1. Introduction to the proposed challenges (projects) and organisation of project teams.
CP2. Information research methodologies for framing the project theme.
CP3. Practical approaches to project development from an ethics-by-design perspective.
CP4. Tools for each stage of project development.
CP5. Usual models for communicating data and results.
Being a project-based course, there is no 100% examination. Assessment runs throughout the semester and consists of the different stages of project development, where:
(i) Each stage is marked by a deliverable (written or presented in class) (E) with feedback.
(ii) An intermediate presentation (A1) with feedback.
(iii) A final presentation (A2) with discussion.
(iv) A poster (informative) (P).
(v) A final project report (R).
The grade will be the result of E x 0.15 + A1 x 0.10 + P x 0.10 + A2 x 0.30 + R x 0.35.
Dependente dos temas específicos do desafio em que cada grupo de estudantes irá desenvolver os trabalhos do projeto.
Voeneky, S., Kellmeyer, P., Mueller, O., & Burgard, W. (Eds.). 2022. The Cambridge Handbook of Responsible Artificial Intelligence: Interdisciplinary Perspectives. Cambridge: Cambridge University Press.
Witten, I. H., Frank, E., Hall, M. A., & Pal, C. J. 2016. Data Mining: Practical machine learning tools and techniques. Morgan Kaufmann.
Provost, F., & Fawcett, T. 2013. Data Science for Business: What you need to know about data mining and data-analytic thinking. O'Reilly Media, Inc.
Web Interfaces for Data Management
After finishing this unit a student should be able to:
LG1. Know and understand basic concepts and technologies for Web development.
LG2. Know and understand interface technologies between a Web application and a Database.
LG3. Model and develop a Web application allowing to manage persistant data from human interaction with software on the Web.
CP1 [Introduction]
- The history of the Web;
- Previous and actual programming languages for the web;
- W3C standards;
CP2 [Modelling and programming a Web application]
- Client-server architecture;
- MVC architecture for the Web.
- Main graphical formatting languages for the Web;
- Libraries for graphical formatting;
- Main programming languages for the Web;
- Libraries for programming for the Web;
- Introduction to security on the client and on the server side.
CP3 [Database access]
- Database access from the Web;
- Data model on the Web application and corresponding interaction with the Database.
CP4 [Data Storage and Management]
- Storage of Web data in a Database;
- Data management.
Given the practical nature of the content taught, assessment will take place throughout the semester, and there will be no exam-based assessment.
First assessment period:
Exercises (10%), completed in class and in groups.
Project (40%), completed in groups, including a report and software. Mandatory, with a minimum grade of 7 out of 20. If there are doubts regarding the project assessment, the teaching team may call the group for an oral discussion.
Test (50%), completed individually. Mandatory, with a minimum grade of 7 out of 20. If the Test grade is more than 5 points lower than the Project grade, the Project grade will be reduced so that the difference between the two grades is at most 5 points.
Second assessment period:
Possibility of retaking the Test (50%) for those who failed the Test taken in the first assessment period.
The grades obtained throughout the semester for the Exercises (10%) and the Project (40%) will be taken into account. Class attendance is not mandatory.
Special assessment period:
Students registered for the special assessment period will complete only the Test (50%) and Project (50%) components.
Improvement of the overall assessment grade may only be carried out in the following academic year.
Mitchell, R. (2016). Web Scraping with Python: Collecting Data from the Modern Web. Ed. O?Reilly Media, Inc. ISBN-13: 978-1491910290. ISBN-10: 1491910291.
Vincent W. S. (2018). Build websites with Python and Django. Ed: Independently published. ISBN-10: 1983172669. ISBN-13: 978-1983172663.
Dean J. (2018). Web Programming with HTML5, CSS, and JavaScript. Ed: Jones & Bartlett Learning. ISBN-13: 978-1284091793. ISBN-10: 1284091791.
Ryan J. (2013). A History of the Internet and the Digital Future. Ed: Reaktion Books. ISBN-13: 978-1780231129
Lambert M. and Jobsen B. (2017). Complete Bootstrap: Responsive Web Development with Bootstrap 4. Ed: Impackt Publishing. ISBN-10: 1788833406. ISBN-13: 978-1788833400.
Downey A. B. (2015). Think Python: How to Think Like a Computer Scientist. Ed: O'Reilly Media. ISBN-10: 1491939362. ISBN-13: 978-1491939369.
Applied Project in Data Science
LO1: Apply methodologies appropriate to the challenge at hand
LO2: Develop skills in cleaning, preprocessing, and integrating real data
LO3: Transform, summarize and visualize data effectively
LO4: Analyze data and interpret results critically
LO5: Communicate results clearly and in a structured manner
CP1: Preliminary data analysis. Exploring libraries for manipulating, cleaning, visualizing, and processing real data, with a focus on preparing data for analysis.
CP2: Variable Analysis and Feature Engineering. Identifying key variables and applying effective selection and transformation techniques to enhance model accuracy and the overall quality of analysis.
CP3: Visualization of univariate and multivariate data.
CP4: Analysis, interpretation, and effective communication of results to support data-driven decision-making.
CP5: Development of an applied project using real-world data, encompassing all stages of the analytical process — from problem definition to the presentation and defense of data-driven solutions
Taking into account the practical nature of the course, assessment is conducted continuously throughout the semester and does not include a final exam. Evaluation is based on two main components:
a) Group quizzes with consultation (40%)
a.1) Two online group quizzes, carried out in class (5% each);
a.2) One comprehensive online group quiz, on the full course content (30%)
b) Group Project (60%)
b.1) Ongoing weekly progress and participation (10%)
b.2) Final written report and oral presentation (50%)
Minimum Requirement: Students must achieve a minimum grade of 9 (out of 20) in each component to pass the course.
(1) Hastie, T., Tibshirani, R., Friedman, J. H., & Friedman, J. H. (2009). The elements of statistical learning: data mining, inference, and prediction (Vol. 2, pp. 1-758). New York: Springer.
https://link.springer.com/book/10.1007/978-0-387-84858-7
(2) James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning: with applications in R (Vol. 103). New York: Springer.
https://link.springer.com/book/10.1007/978-1-0716-1418-1
(3) Cady, F. (2024). The data science handbook. John Wiley & Sons.
https://www.wiley.com/en-us/The+Data+Science+Handbook%2C+2nd+Edition-p-9781394234493
(4) Boehmke, B., & Greenwell, B. M. (2019). Hands-on machine learning with R. Chapman and Hall/CRC.
https://bradleyboehmke.github.io/HOML/, https://bradleyboehmke.github.io/HOML/
(5) Wes McKinney (2022), Python for Data Analysis, 3rd Edition, O'Reilly Media, Inc. (https://wesmckinney.com/book/, https://github.com/wesm/pydata-book)
(1) Harrison, M., & Petrou, T. (2020). Pandas 1. x Cookbook. Packt Publishing.
(2) Mukhiya, S. K., & Ahmed, U. (2020). Hands-On Exploratory Data Analysis with Python: Perform EDA techniques to understand, summarize, and investigate your data. Packt Publishing Ltd.
https://github.com/PacktPublishing/hands-on-exploratory-data-analysis-with-python
(3) Gagolewski, M. (2022). Minimalist Data Wrangling with Python. arXiv preprint arXiv:2211.04630.
https://datawranglingpy.gagolewski.com/
(4) Vasconcelos, J. B., & Barão, A. (2017). Ciência dos dados nas organizações. FCA Editora.
Symbolic Artificial Intelligence for Data Science
The course introduces the major themes of (mostly) Symbolic Artificial Intelligence and Machine Learning, from an essentially applied perspective, bearing in mind the major context provided by the data science degree, the knowledge and skills acquired in the other courses, and the fundamental objectives and requirements of the data science degree.
The three major topics of the program are logic programming, mostly symbolic adaptive techniques for the representation of adaptive world models, and symbolic machine learning algorithms to learn world models.
After the students have completed the course, they must
? Be fully aware of the existence of mainly symbolic paradigms for the representation and autonomously learning of adaptive world models.
? Have mastered the capability to decide whether to use the paradigms learned in the course to application problems / domains whenever suited.
Overview of the Curricular Unit: the need, advantages and disadvantages of essentially symbolic technologies for representing and learning adaptive models of reality, and the role of each programme component in the desiderata of the chair.
Programming in logic to represent models of reality and to reason with them.
Representation and reasoning based on fuzzy sets and in fuzzy logic to represent essentially symbolic adaptive models and reason with them.
Representation and reasoning based on cases to represent essentially symbolic adaptive models and reason with them.
Introduction to Explainable AI and its characteristics and application domains.
Concepts of Responsible AI.
In semester assessment, students will have to take:
- Individual written test on the entire CU programme (60%) - occurring during exams' period (1st or 2nd exam).
- (Group) research work on one of the CU topics, with a report and an oral presentation (40%). The oral presentation is done in class time during the semester. The grade of the research work is split 50% for each item and the members of the group may have different grades.
Both assessment components on semester evaluations have a minimum mark of 8.
Alternatively, students can take only one exam (100%), which can be at both dates of exams.
At special exams' period the students take the exam (100%).
Whenever there are doubts regarding the authorship or mastery of the knowledge demonstrated in an assessment component, the student may be required to take an individual oral or written exam to validate the corresponding grade.
Logic Programming and Inductive Logic Programming:
Ivan Bratko. 2011. Prolog Programming for Artificial Intelligence (4th Edition). Pearson Education Canada (International Computer Science Series).
Fuzzy Systems:
Guanrong Chen, and Trung Tat Pham. 2005. Introduction to Fuzzy Systems. CRC Press.
Case based reasoning:
Michael M. Richter, and Rosina Weber. 2013. Case-Based Reasoning. A Textbook. Springer-Verlag Berlin Heidelberg
Lynne Billard, Edwin Diday. 2007. Symbolic Data Analysis: Conceptual Statistics and Data Mining, John Wiley & Sons, Ltd, Chichester, UK
Longitudinal Models
Upon successfully completing this course, the student is expected to be able to:
LO1. Understand the structure and characteristics of longitudinal data and repeated measures.
LO2. Understand the most common time series models.
LO3. Apply and interpret mixed-effects regression models (linear and non-linear models).
LO4. Understand growth modeling and change in longitudinal data.
LO5. Analyze longitudinal data using computational tools such as R and Python.
LO6. Develop a critical approach to selecting appropriate models and interpreting the results.
S1. Introduction to Longitudinal Data
S2. Time Series
S2.1. Lag operator, stationarity, unit root test
S2.2. White noise, ARMA, ARIMA, SARIMAX models
S2.3. Box-Jenkins methodology, forecasting
S2.4. Applications with R/Python
S3. Panel Data Models
S3.1. Fixed effects models
S3.2. Mixed effects models
S3.3. Growth and change models
S3.4. Generalized linear models
S3.5. Covariance structure in longitudinal data
S3.6. Applications with R/Python
S4. Case Studies
Assessment throughout the semester includes:
a) Group work weighting 40%, with the possibility of discussion if teacher considers it necessary. The minimum grade for the work is 10 points (out of 20). A group presentation should be made in class, and a final report should be handed in, as explained in the Guidelines to the Group Coursework presented to the students.
b) Two individual tests, weighting 20 and 40% of the final mark, the first conducted in October and the second after the end of the classes. A minimum grade of 9.5 points is required in both tests for students to remain in assessment throughout the semester. To approve a weighted average of at least 9.5 points out of 20 is required.
None of these marks can be used in the exam assessment.
Assessment by exam:
An individual exam covering the entire syllabus, with a minimum grade of 10 (grade rounded to the nearest unit).
Both the individual test and the exam will be carried out without consulting support sheets, books, or other materials, and the use of graphing calculators or smartphones is not permitted; Students can only consult the form and tables made available in Moodle for this purpose.
The Course Unit Coordinator may require a supplementary oral discussion regarding any assessment component or instrument specified in this Course Unit Outline whenever this is deemed necessary to verify the authorship of submitted work, confirm the acquisition of the knowledge and competences being assessed, or clarify discrepancies between different assessment components.
The supplementary oral discussion does not constitute an independent assessment instrument, carries no separate weighting in the final grade for the course unit, and is intended solely to validate the grade awarded for the corresponding assessment component or instrument. The grade previously assigned to that component or instrument is considered provisional until the supplementary oral discussion has taken place and may be confirmed or revised in light of the validation process and the applicable assessment criteria.
The student will be notified of the supplementary oral discussion at least 24 hours in advance, with the date and time of the discussion specified. Failure to attend, without a valid justification, will make it impossible to validate the relevant assessment component or instrument, and the consequences applicable to failure to complete or withdrawal from that assessment component will apply.
(1). Benjamin E. Schlegel, (2026), Data Science for the Social Sciences: Introduction and Advanced Models with R, Springer.
(2). Manu Joseph, (2022), Modern Time Series Forecasting with Python, Packt.
(3). Baltagi, B. H. (2021) Econometric Analysis of Panel Data, 5th ed. Wiley. (with Student Companion Site)
(1). Diana Mendes, (2026), Slides, scripts e notebooks, Moodle
(2). Xian Liu, (2016), Methods and Applications of Longitudinal Data Analysis, Elsevier.
(3). Yves Croissant and Giovanni Millo, (2019), Panel Data Econometrics with R, Wiley.
Programme Structure for 2026/2027
Linear Algebra Fundamentals
By the end of the course, each student should be able to:
OA1. Define vectors and explain their properties. Perform operations with vectors in Euclidean space R^n. Define and determine subspaces of R^n, their bases, and dimension.
OA2. Apply the methods of Gauss and Gauss-Jordan elimination to solve and classify linear systems. Interpret their solutions geometrically.
OA3. Give examples of different types of matrices and perform operations with matrices. Formulate relationships between matrices, vectors, and linear systems.
OA4. Recognize a linear transformation. Determine the associated matrices, kernel, and image subspaces. Perform basis changes.
OA5. Calculate determinants. Explain their properties and applications.
OA6. Define and determine eigenvalues and eigenvectors. Diagonalize matrices. Calculate integer powers of diagonalizable matrices.
CP1. Vectors
The vector space R^n. Inner product and norm. Linear combinations and linear independence. Bases and dimension. Coordinates.
CP2. Systems of Linear Equations
Gaussian elimination method. Classification of linear systems.
CP3. Matrices
Addition and scalar multiplication. Matrix multiplication. Transposition. Inverse matrix and properties.
CP4. Linear Functions
Linear function. Matrix of a linear function. Kernel and image subspaces and the dimension theorem. Basis change.
CP5. Determinants
Definition and properties of the determinant. Determinants and elementary operations.
CP6. Eigenvalues and Eigenvectors
Eigenvalues and eigenvectors. Eigen subspaces. Diagonalization.
Students may choose one of the following assessment methods:
• Assessment throughout the semester:
o 4 in-class mini-tests (20%): written tests performed in class, each lasting 15 minutes;
o Written exam (80%): written exam taken during the 1st exam period, with a minimum grade of 8.5 out of 20.
Students under assessment throughout the semester will receive a grade of zero for any mini-tests not taken.
Mini-tests must be taken in the class group in which the student is enrolled.
mini-tests will be administered either at the beginning or end of the scheduled class, and the rest of the class will proceed as usual.
For the purpose of calculating the final grade for this component, the best 3 out of the 4 mini-tests scores will be considered.
For students choosing this method, the final grade will be the higher of the two: the assessment throughout the semester or the final exam assessment.
• Final exam assessment:
Students will take a written exam worth 100% of the final grade, during either the 1st or 2nd exam period.
The minimum passing grade for this course unit is 10 out of 20.
The course instructors reserve the right to conduct oral exams when necessary.
Strang, G. (2023). Introduction to Linear Algebra (sixth edition) Wellesley-Cambridge Press.
Apoio teórico fornecido pelos docentes. Caderno de exercícios fornecido pelos docentes.
Lay, D., Lay, S., & McDonald, J. (2016) Linear Algebra and Its Applications (fifth edition) Pearson.
Calculus Topics I
At the end of this course the student should be able to:
LO1: Calculate limits of sequences
LO2. Compute derivatives and interpret the corresponding result.
LO3. Determine linear and higher order approximations.
LO4. Explicitly compute the antiderivative of some elementary functions.
LO5. Use the fundamental theorem of calculus to differentiate functions defined by integrals.
LO6. Use integrals to compute areas, lengths, probabilities, etc.
LO7. Integrate some notable ODEs.
LO8. Apply some simple numerical methods to compute approximate values and use graphical computational tools.
1. Sequences
1.1. Some concepts
1.2. Convergence
1.3. Some limits and useful results
2. Differential calculus in R
2.1. A brief review
2.2. Continuity and limits
2.3. Differentiability and Taylor’s formula; applications
2.4. Numerical methods
2.4.1. Fixed point method
2.4.2. Bisection method
2.4.3. Newton-Raphson method
2.4.4. Numerical differentiation
3. Integral calculus in R
3.1. Anti-derivatives
3.2. Integrals
3.3. Fundamental theorem of calculus
3.4. Numerical integration
3.4.1. Mid-point method
3.4.2. Trapezoidal rule
4. Ordinary differential equations.
4.1. Separable variables
4.2. First order linear equations
4.3. Numerical methods
4.3.1. Euler’s method
4.3.2. Runge-Kutta method (RK4)
A student must obtain an overall grade of at least 10 (out of 20) in one of the following assessment modes:
- Assessment during the semester: a mid-term test (37.5%) + final test (37.5%) + two group works, one about numerical calculus and the other about graphical representation (25%, 12.5% each).
- Exam assessment: in any of the exam seasons (100%).
The exam consists of two parts: analytical (75%) and numerical (25%). Students who have successfully completed the teamwork may skip this numerical component.
The minimum grade is 8. Students with a grade over 16 should be submitted to an oral examination.
[1] Ferreira, J.C. (2011). “Introdução à Análise Matemática”, Fundação Calouste Gulbenkian.
[2] Strang, G. (1991). “Calculus”, Wellesley-Cambridge.
[3] Caputo, H.P. (1973). “Iniciação ao Estudo das Equações Diferenciais”, Livros Técnicos e Científicos Editora, S.A.
[4] Suleman, A., Rocha, J., Alho, A., Apontamentos de aula. (disponível no Moodle)
[5] Suleman, A., Notas sobre cálculo numérico (disponível no Moodle).
[6] Santos, M.I.R., Matemática computacional (IST).
Optimization for Data Science
By successfully completing this Curricular Unit, students will be able to:
LO1. Develop formulations in linear programming, integer linear programming, and non-linear programming for efficiently solve complex problems in real contexts.
LO2. Interpret outputs obtained by general software for solving problems formulated in linear programming, integer linear programming, and non-linear programming.
LO3. Do the economic interpretation and produce recommendations based on the results obtained for problems formulated in linear programming, integer linear programming, and non-linear programming.
Programmatic Contents (PC):
PC1: Linear Programming
1.1 General form of a Linear Programming model
1.2 Formulating problems in Linear Programming
1.3 Graphical resolution
1.4 Resolution using general software (Excel Solver or other)
1.5 Interpreting results and sensitivity analysis
PC2: Integer Linear Programming
2.1 Formulating problems in Integer Linear Programming
2.2 Formulating problems with binary variables
2.3 Resolution using general software (Excel Solver or other)
2.4 Interpreting results
2.5 Branch-and-Bound algorithm
PC3: Non-linear Programming
3.1 Formulating problems in Nonlinear Programming
3.2 Resolution using general software (Excel Solver or other)
3.3 Interpreting results
Assessment throughout the semester or assessment by exam:
1. Assessment throughout the semester:
a) 5 individual short tests: i) weighting 10%; ii) completed in class throughout the lecturing period; iii) grade for this component: arithmetic mean of the 4 best grades obtained on the 5 short tests;
b) Group project: i) weighting 30%; ii) Groups of 4 students; iii) with oral discussion;
c) Individual final written test: i) weighting 60%; ii) minimum classification of 8.5;
d) Attendance of at least 2/3 of the classes taught;
e) Approval: minimum weighted average of 9.5.
2. Evaluation by Exam (1st and 2nd Season):
a) Individual written exam: i) weight of 100%; ii) approval: minimum classification of 9.5.
In both assessment methods, students may be required to take an individual exam, either oral or written, aimed at validating the grade obtained.
Scale: 0-20 points.
* Ragsdale, C.T. (2017). Spreadsheet Modeling and Decision Analysis: A Practical Introduction to Business Analytics. 8th Ed. Cengage Learning.
* Evans, J. (2021). Business Analytics. 3rd Ed. Global Edition. Pearson.
* Hillier, F.S and Lieberman, G.J. (2015). Introduction to Operations Research, 10th Ed., McGraw-Hill.
* Ragsdale, C.T. (2001). Spreadsheet Modeling & Decision Analysis: A Practical Introduction to Management Science. 3rd Ed., South-Western College Publishing.
* Wolsey, L.A. (1998). Integer Programming. Wiley.
Probabilities and Sampling
At the end of the curricular unit, students should be able to simulate probabilistic scenarios (LO1), calculate conditioned probabilities (directly or via Bayes' theorem), and verify the independence of events (LO2). They must know the
classical statistical sampling methods and their applicability conditions (LO3) and know how to calculate estimates and their precision measures (LO4). They should be able to identify contexts of complementarity between survey sampling and big data (LO5).
CP1. Probability theory: definitions, axioms, conditional probability, total probability and Bayes' formula.
CP2. Univariate random variables: probability and density function, distribution function, parameters.
CP3. Survey methodology: data collection by questionnaire; sampling plan.
CP4. Random vs. Non-random sampling: types of non-random sampling.
CP5. Random sampling: simple, systematic, with unequal probabilities, stratified.
CP6. Combining surveys with big data.
Assessment regime: throughout the semester or by exam.
Throughout the semester:
- Mandatory attendance of at least 75% of the classes (18 classes).
- Midterm test (50%) + final test (50%).
- Both tests have a minimum grade of 7.5 values and consist of two components: theoretical-interpretive (70%) and applied assessment with R (30%).
By exam:
- Exam (100%) consists of two components: theoretical-interpretive (70%) and applied assessment with R (30%).
Whenever there are doubts regarding the authorship or level of the knowledge demonstrated in an assessment component, the student may be required to take an individual oral or written test to validate the corresponding grade. An oral exam may also be required if the grade is 18 or higher; if the student does not attend the oral exam, the final grade will be 16 values.
The use of cell phones, smartwatches, or any other electronic devices is not permitted during written exams. Unauthorized use of such devices will be penalized in accordance with the Student Disciplinary Regulations.
Groves, R., Fowler, F., Couper, M., Lepowski, J., Singer, E. & Tourangeau, R. (2009) Survey Methodology, 2nd edition, John Wiley and Sons.
Levy, P., Lemeshow, S. (1991). Sampling of Populations: methods and applications. Wiley Interscience.
Reis, E., Andrade, M., Calapez, T. & Melo, P., Estatística Aplicada, volume 1. 7ª edição. Lisboa. Edições Sílabo, 2021, ISBN 978-989-561-186-7.
Reis, E., Andrade, M., Calapez, T. & Melo, P., Estatística Aplicada volume 2, 6ª edição, Lisboa. Edições Sílabo, 2019, ISBN 978-972-618-986-2.
Speegle, D., & Clair, B. (2021). Probability, Statistics, and Data: A Fresh Approach Using R. Chapman and Hall/CRC. Free access at https://mathstat.slu.edu/~speegled/_book/
Salganik, M. (2018). Bit by Bit- Social Research in the Digital Age. New Jersey: Princeton University Press Verzani, J. (2014). Using R for Introductory Statistics, 2nd Edition, Chapman & Hall/CRC, eBook ISBN 9781315373089. https://cran.r-project.org/doc/contrib/Verzani-SimpleR.pdf
Calculus Topics II
At the end of the course, each student should be able to:
LG1. Compute partial derivatives and gradients.
LG2. Determine linear approximations of functions of several variables.
LG3. Determine and classify critical points of functions of several variables.
LG4. Apply the previous concepts in the context of regression problems.
LG5. Compute double integrals.
LG6. Apply integral calculus to the evaluation of volume, mass and probability.
LG7. Interpret geometrically all the previous concepts.
LG8. Implement in MATLAB some of the computacional methods studied in class.
PC1. Differential calculus
1.1. Limits and continuity
1.2. Partial derivatives.
1.3. Tangent plane and differentiability.
1.4. The chain rule
1.5. Computation and classification of critical points.
1.6. Gradient descent.
1.7. Linear regression.
PC2. Integral calculus.
2.1. Double integral.
2.2. Double integrals in polar coordinates.
2.3. Application of integral calculus to the evaluation of volume, mass and probability.
Students must obtain an overall grade of at least 10 (out of 20) in one of the assessment modes:
- Assessment throughout the semester: 2 Written Tests (40% each) + participation in class (20%).
- A final Exam (100%) in either the 1st or 2nd examination period.
Stewart, J. "Cálculo - Volume 2", Tradução da 8ª edição norte-americana (4ª edição brasileira), Cenage Learning, 2017.
Data in Science, Bussiness and Society
After the course the student should be able to achieve the Learning Outcomes (LO):
OA1: Account for different definitions of data, different data types and different research approaches that generate it.
OA2: Identify the knowledge claims underlying different interpretations of data.
OA3: Explain the difference between quantitative and qualitative approaches to data generation.
OA4: Examine the implications of data collection for research, business and society.
OA5: Discuss different debates about the implications of data for people in organizations and society.
CP1: What data are and how to think with data.
CP2: Types of problems addressed in Data Science and specificities in the domains of Science, Management, and Society.
CP3: Different traditions and research methodologies and definitions of knowledge acquisition.
CP4: Translating real challenges into technical concepts and using scientifically oriented language.
CP5: The ethical dimension of data use strategies.
CP6: Presentation of practical cases.
This course uses only assessment throughout the semester and does not include exams.
Assessment components:
a) Mini-tests (30%): 6 mini-tests (5% each, the vast majority to be taken at home)
b) Project (30%): group assignment
c) Final test (40%): Written test to be taken during the 1st season, 2nd season or special season (Art. 14, RGACC)
Passing requirement: Final test >= 8 points (out of 20 points)
The final grade for the Project will depend on the code, the reports, and the student's performance in presenting their work.
Cathy O'Neil, Rachel Schutt, Doing Data Science: Straight Talk from the Frontline, 2014, ISBN: 9781449358655,
Borgman, C. L., Big data, little data, no data: scholarship in the networked world, 2015, ISBN: 9780262529914,
Rob Kitchin, The data revolution: Big data, open data, data infrastructures and their consequences, 2014, https://doi.org/10.4135/9781473909472,
Davenport, T., Harris, J., and Morison, R., Analytics at work: smarter decisions, better results. Harvard Business Review Press, USA., 2010, ISBN: 9781422177693,
Turban, E., Sharda, R., Delen, D., Decision Support and Business Intelligence Systems (9th Eds), 2010, ISBN: 978-0136107293,
Davenport, T., Big Data at Work: Dispelling the Myths, Uncovering the Opportunities, 2014, ISBN: 978-1422168165,
Programming
After obtaining approval in the course, students should be able to:
OA1. Develop functions/procedures that implement simple algorithms.
OA2. Develop code that manipulates arrays and objects.
OA3. Develop simple object classes, considering the notion of encapsulation.
OA4. Write and understand Python code.
CP1. Variables and control structures
CP2. Functions and parameters
CP3. Invocation and recursion
CP4. Procedures and input/output
CP5. Objects and references
CP6. Object classes
CP7. Composite objects
CP8. Composite object classes
CP9. Arrays
CP10. Matrices
This course is done only by assessment throughout the semester, not considering the modality of assessment by exam.
Evaluation components:
a) HWs (15%): 6 online mini-tests, to do at home;
b) TEST1 (20%): Closed-book written test during the intermediate evaluation week;
c) PROJECT (25%): Individual project;
d) TEST2 (40%): Closed-book written test during the 1st season, 2nd season or special season (Art. 14 RGACC).
Approval requirement: TEST2 >= 8 points.
The final grade for the PROJECT is determined for each student through an oral exam or practical test, which uses the project submitted by the student as a basis to assess the acquired knowledge.
Other relevant information:
- Attendance is not an essential requirement for approval
- Questions asked in the written tests may involve aspects related to the project;
- Whenever there are doubts regarding the authorship or the mastery of the knowledge demonstrated in an assessment component, the student may be summoned to an individual oral or written exam intended to validate the respective grade;
- It is not possible to pass only by taking the final exam;
- in case of failure in the 1st season, the student can take TEST2 in the 2nd season, keeping the grade of the other components;
- When the grade improvement occurs in a school year different from the one in which the work was done, the grade of the components PROJECT, HWs and TEST1 is replaced by a practical test, to be performed on a computer before or after the written exam. Students under these conditions who wish to improve their grades should contact the course coordinator in advance, at least 2 days before the 1st season.
João P. Martins, Programação em Python: Introdução à programação com múltiplos paradigmas, 2013, IST Press, https://istpress.tecnico.ulisboa.pt/produto/programacao-em-python-introducao-a-programacao-utilizando-multiplos-paradigmas/
Exploratory Data Analysis
Learning goals (LG) to be developed in articulation with the general objectives:
LG1. Prepare data for analysis.
LG2. Use and interpret a set of statistical tools in the field of descriptive.
LG3. Use Excel, R and Jamovi in data preparation, analysis and representation applications.
LG4. Adapt the visual representation models to different objectives, according to good visualization practices.
LG5. Interpreting and writing the results of a descriptive data analysis.
Syllabus contents (SC) articulated with the learning objectives.
SC1. Organization, preparation and transformation of data
SC2. Exploratory data analysis
Missing values
Coding and imputation
Exploratory charts
Random variables
Empirical distribution function
Normal Distribution
SC3. Descriptive data analysis
Descriptive measures
Single and bivariate analysis
Association measures
SC4. Visual representation
Introduction to the principles of visual representation
Visual representation structures
Assessment throughout the semester:
- Individual exercise with R (10%)
- 3 mini-tests to be taken during lessons, with the two highest marks being considered for the weighting in the overall assessment (10% + 10%)
- Group work, with presentation and discussion (30%); minimum grade 7.5
- Written test (40%); min. grade 7.5
A minimum attendance of 70% of class hours is required for assessment throughout the semester.
Assessment by exam:
- Exam/individual practical work (35%); minimum grade 7.5
- Written exam (65%); minimum grade 7.5
Brown, D.S. (2022). Statistics and Data Visualization Using R. The Art and Practice of Data Analysis. Sage Publication, Inc.
Cairo, A. (2013). The Functional Art: An introduction to information graphics and visualization (Voices That Matter).
New Riders.
Carvalho, A. (2017). Métodos quantitativos com Excel, Lisboa, Lidel edições técnicas.
Chang, W. (2024) R Graphics Cookbook. 2nd ed. O’Reilly. (Disponível em: https://r-graphics.org/)
Reis, E. (1998). Estatística Descritiva, Lisboa, Sílabo,7ª ed.
Rocha, M. & Ferreira, P.G. (2017) Análise e Exploração de Dados com R. Lisboa, FCA
Alexandrino da Silva, A. (2006). Gráficos e mapas-representação de informação estatística. Lisboa, Lidel edições técnicas.
Barroso, M., Sampaio, E. & Ramos, M. (2003). Exercícios de Estatística Descritiva para as Ciências Sociais, Lisboa, Sílabo.
Carvalho, A. (2017). Gráficos com Excel - 95 Exercícios, Lisboa, FCA.
Dias Curto, J.J., & Gameiro, F. (2016). Excel para Economia e Gestão. Lisboa, Ed. Sílabo.
Field, A., Miles, J., & Field, Z. (2012). Discovering statistics using R. London, SAGE Publications Ltd.
Hoaglin, D.C., Mosteller, F & Tukey, J. W. (1992). Análise Exploratória de Dados. Técnicas Robustas, Ed. Salamandra, Lisboa.
Murteira, B. J. F. (1990). Análise Exploratória de Dados. Estatística Descritiva, McGraw Hill.
Wickham, H. (2015). ggplot2: Elegant Graphics for Data Analysis, Springer. (Disponível em: https://ggplot2-book.org/)
Data Structures and Algorithms
At the end of this course, students should be able:
LO1: Identify, rewrite, and examine common forms of data organization and its associated algorithms (with and without dynamic memory management, with iterative or recursive algorithms);
LO2: Identify the most appropriate and efficient data structure for a problem;
LO3: Know how to evaluate and compare the order of performance and efficiency of a given algorithm and/or data structure for the common operations of inserting, removing, and accessing;
LO4: Understand the pros and cons of recursive, and iterative algorithms, as well as dynamic programming.
LO5: Understand different search and sorting algorithms appropriate for computational solutions.
CP1: Data Structures and Algorithms: what are these and why are they important? Abstract Data Types
CP2: Linear data structures: stacks, queues, double-ended queues, lists, and linked lists.
CP3: Introduction to algorithm complexity analysis.
CP4: Search algorithms: linear, and binary search.
CP5: Recursion, iteration, backtracking, and dynamic programming.
CP6: Iterative sorting algorithms: Bubblesort, Selectionsort, and Insertionsort.
CP7. Recursive sorting algorithms: Mergesort, and Quicksort.
CP8: Nonlinear data structures: tree, binary search trees, AVL trees, and graphs.
CP9: Simple algorithms for nonlinear data structures.
It has been approved by the ISTA Pedagogical Committee that the EDA assessment is done only through assessments throughout the semester (see Observations 1). Therefore, it is not possible to assess EDA with an exam worth 100%.
Minimum attendance criteria: 75% of the 12 weeks x 3 hours of class.
Assessment elements and their respective weightings in the final grade:
- Individual written test -> 60%, minimum grade of 7.5 points, plaaned to be held in the after-work class of the last week of classes;
- Individual assignment 1, with oral exam -> 15%;
- Individual assignment 2, or in groups of 2 (to be decided), with individual oral exam -> 25%, minimum grade of 7.5 points.
The final grade is obtained with the weighted average: Final_grade = 60% x Test_grade + 15% x Assignment_1_grade + 25% x Assignment_2_grade.
The classifications of the 3 assessment elements are used in the formula without rounding to the nearest whole number. To pass the course, the final grade must be greater than or equal to 10 points.
Students must be familiar with the Student Disciplinary Regulations (see Observations 2) and fulfill their general duties, as described in point 2 of Article 2 of those regulations (see Observations 2).
Students with special status conferred by the Serviços de Gestão do Ensino (contact these Services, as professors are not competent to grant this status) may have access to the special period (see Observations 3). In the Special Period, the assessment elements and their respective weightings in the final grade are:
- individual written exam -> 60%, minimum grade of 7.5 points;
- two individual assignments, with an individual oral exam -> 15% + 25%, minimum grade of 7.5 in each of the assignments.
The final grade in the special period is obtained in the same way as in the regular period (assessment throughout the semester).
- J. Wengrow, A Common-Sense Guide to Data Structures and Algorithms, Second Edition. The Pragmatic Bookshelf, 2020.
- M. Goodrich, R. Tamassia, and M. Goldwasser, Data Structures & Algorithms in Python. Wiley, 2013.
- B. Miller and D. Ranum, Problem Solving with Algorithms and Data Structures using Python, Second Edition, Release 3.0. 2013.
- T. Cormen, C. Leiserson, R. Rivest, and C. Stein, Introduction to Algorithms, Fourth Edition. MIT Press, 2022.
- Referências adicionais a indicar durante as aulas.
Big Data Storage
LO1 - Implement data storage solutions in distributed and fault-tolerant support;
LO2 - Manipulation and extraction of large amounts of information from unstructured databases;
LO3 - Develop soft skills, namely problem solving, teamwork and collaboration, and critical observation.
CP1 - Introduction to Non-Relational Databases;
CP2 - Redundancy to manage fault tolerance;
CP3 - Data distribution to manage large volumes of information;
CP4 - Introduction to MongoDB;.
CP5 - Database Design in MongoDB;
CP6 - Manipulation of JSON structures;
CP7 - Data extraction in MongoDB.
Assessment throughout the semester is done through a written test (minimum grade: 7.5) which takes place on the same date as the 1st season exam and which is worth 60% of the grade and a group work, 40% of the grade (minimum grade: 7.5), to be delivered in the last week of classes.
Alternatively, there is assessment by exam - 100% of the grade (1st season, 2nd season and special season).
NoSQL Database: New Era of Databases for Big data Analytics - Classification, Characteristics and Comparison, A B M Moniruzzaman, Syed Akhter Hossain, 2013 (https://arxiv.org/abs/1307.0191)
MongoDb Homepage
Big Data Processing
At the end of the course, students should be able to:
OA1: understand and know the main platforms for processing large amounts of information
OA2: understand and know how to apply distributed programming/computing models
OA3: understand the stages associated with a machine learning project for large amounts of information
OA4: know how to apply dimensionality reduction techniques
OA5: apply supervised or unsupervised learning techniques to large-scale problems
CP1: Computing platforms for big data
CP2: Machine learning pipeline for big data
CP3: Dimensionality reduction
CP4: Supervised/unsupervised learning for large scale
CP5: Case studies: PageRank and Recommendation Systems
This course includes the following assessment methods: (1) assessment throughout the semester; (2) assessment by exam.
(1) Assessment throughout the semester
The final grade is made up of:
- Individual written test (70%), with a minimum mark of 8.0;
- Group work (30%).
The group work has a mid-term submission, which will count for 30%, and a submission at the end of the semester, which will count for 70%. Those who do not submit the mid-term portion will automatically be assessed by exam.
The work will include an oral presentation/discussion, and the final grade will be individual.
(2) assessment by exam
The final grade will be based on a single written exam, including special season.
- Practical Data Science with Hadoop and Spark: Designing and Building Effective Analytics at Scale, Ofer Mendelevitch, Casey Stella and Douglas Eadline, Addison-wesley, 2016.
- Advanced Analytics with Spark: Patterns for Learning from Data at Scale, Sandy Ryza et al., O'Reilly Media, 2017.
- Learning Spark: Lightning-Fast Big Data Analysis, Holden Karau, A. Konwinski, P. Wendell and M. Zaharia, O'Reilly Media, 2015.
- Big Data: Algorithms, Analytics, and Applications, Kuan-Ching Li et al., Chapman and Hall/CRC, 2015.
- Mining of Massive Datasets, A. Rajaraman, J. Ullman, 2011, Cambridge University Press.
- The elements of statistical learning, Trevor Hastie, Robert Tibshirani, and Jerome Friedman. Springer, 2001
- All of Statistics: A concise course in Statistical Inference, L.Wasserman, Springer, 2003.
Security, Ethics and Privacy
LG1. Recognize the main security issues in software-based systems, their causes, and consequences.
LG2. Identify and describe the security services necessary to implement a specific information protection policy based on risk analysis.
LG3. Learn the principles and regulatory frameworks in the domains of personal data protection and privacy, with special focus on the General Data Protection Regulation of 2016.
LG4. Ethically and critically reflect on the implications of technologies and data processing on individuals and society, addressing the resulting challenges in the fields of information security, data protection, and privacy.
CP1. Information Security:
Fundamentals of security – data security;
Vulnerabilities and threats in security;
IRM – Information Risk Management;
Cryptography – symmetric, asymmetric, hash functions, MAC (Message Authentication Code), digital signatures;
Public Key Infrastructures (PKI).
CP2. Privacy and Personal Data Protection:
The GDPR and Law 58/2019;
Anonymization and pseudonymization techniques.
CP3. Ethics:
Normative ethical theories;
Regulatory dilemmas and gaps in ICT;
Case studies in applied ethics;
The regulation of Artificial Intelligence
The course unit provides for two assessment modes: assessment throughout the semester and assessment by examination.
Assessment throughout the semester:
Assessment throughout the semester includes:
* Individual tests: throughout the semester, students will take short individual tests, lasting approximately 15–25 minutes each, administered in person and covering material from previous classes. The number of tests may vary depending on the pace of the classes, with these tests accounting collectively for 66% of the final grade.
* A group assignment, carried out within the Ethics module, accounting for 34% of the final grade.
A minimum grade of 7 out of 20 is required in the group assignment. The assignment is assessed on the basis of the content, clarity and consistency of the analysis.
Group assignments with a provisional grade of 17 out of 20 or above are subject to a group oral discussion. The discussion forms part of the assessment of the assignment, and the grade may be adjusted for the group as a whole or individually.
Assessment by examination:
Assessment by examination consists of an individual written examination, accounting for 100% of the final grade, to be taken during the official assessment period and covering all content taught in the course unit. Where necessary, it may include a supplementary oral examination.
In the first examination period, students may choose between assessment throughout the semester and assessment by examination. Students who choose assessment throughout the semester and do not pass may only take the examination in the second examination period.
Use of artificial intelligence:
In the group assignment, the use of artificial intelligence tools is governed by Iscte’s Guide for the Responsible Use of Artificial Intelligence in Academic Work.
Whenever AI tools are used, their use must be declared in accordance with the Guide and accompanied by the applicable institutional form. Students remain fully responsible for the content submitted.
Andress, J. (2014). The Basics of Information Security: Understanding the Fundamentals of InfoSec in Theory and Practice. Syngress.
Kim, D., Solomon, M. (2016). Fundamentals of Information Systems Security. Jones & Bartlett Learning.
Cannon, J.C. Privacy in Technology: Standards and Practices for Engineers and Security and IT Professionals. Portsmouth: AN IAPP Publication, 2014.
Breaux, Travis. Introduction to IT Privacy: A Handbook for Technologists. Portsmouth: An IAPP Publication, 2014.
Whitman, M., & Mattord, H. (2013). Management of information security. Nelson Education.
Katz, J., & Lindell, Y. (2014). Introduction to modern cryptography. CRC press.
Ethics, Technology, and Engineering: An Introduction (2023), Ivovan de Poel, Lamber Royakkers, Wiley-Blackwell, 2nd edition
European Union Agency for Fundamental Rights, The Handbook on European data protection law, 2019, http://fra.europa.eu/sites/default/files/fra_uploads/fra-coe-edps-2018-handbook-data-protection_en.pdf
Floridi, L. (2019). The Logic of Information: A Theory of Philosophy as Conceptual Design. Oxford University Press.
Canto Moniz, G. (2023). Manual de introdução à proteção de dados pessoais. Coimbra: Almedina.
Whitman, M., Mattord, H. (2017). Principles of Information Security. Course Technology.
Bowman, Courtney. The Architecture of Privacy: On Engineering Technologies that Can Deliver Trustworthy Safeguards. O?Reilly Media, 2015.
Anderson, R. J. (2010). Security engineering: a guide to building dependable distributed systems. John Wiley & Sons.
Zúquete, A. (2018). Segurança em redes informáticas. FCA-Editora de Informática.
Regulamentos e orientações da Comissão Europeia relativos à Proteção de Dados, https://www.edpb.europa.eu/our-work-tools/general-guidance/guidelines-recommendations-best-practices_pt
Bynum, Terrell Ward, and Simon Rogerson, (2004), Computer Ethics and Professional Responsibility: Introductory Text and Readings. Oxford: Blackwell, 2004.
Grupo do Artigo 29, Parecer 05/2014 sobre técnicas de anonimização do grupo de trabalho de proteção de dados do artigo 29.º, de 10 de Abril de 2014, 2014, -, https://ec.europa.eu/justice/article-29/documentation/opinion-recommendation/files/2014/wp216_pt.pdf
Enisa, Orientações da Enisa sobre técnicas de pseudonimização e boas práticas, 2019, -, https://www.enisa.europa.eu/publications/pseudonymisation-techniques-and-best-practices
UE, Proposta do regulamento do parlamento europeu e do conselho que estabelece regras harmonizadas em matéria de inteligência artificial (regulamento inteligência artificial) e altera determinados atos legislativos da União, 2023, -, https://eur-lex.europa.eu/legal-content/PT/TXT/?uri=CELEX%3A52021PC0206
Outros textos a indicar e distribuídos pelo docente ao longo do semestre.
Heuristic Optimization
At the end of the Curricular Unit, the student is expected to be able to:
LO1- Discuss challenges faced in real, large scale optimization problems
LO2 - Explain and discuss the available methodologies for addressing hard optimization problems
LO3 - Formulate and design effective solution methods for addressing optimization problems
LO4- Employ the use of advanced tools to solve optimization problems
Programmatic contents (PC):
PC1. MULTIOBJECTIVE PROGRAMMING
1.1. Basic concepts
1.2. Methodologies
1.2.1. Non-Preemptive Goal Programming
1.2.2. Preemptive Goal Programming
PC2. METAHEURISTICS
2.1. Concepts and terminology
2.2. Single point algorithms
2.2.1. Tabu Search
2.2.2. Simulated Annealing
2.3. Genetic Algorithms
1st SEASON ASSESSMENT
In the 1st Season, the Course can be assessed using the Assessment throughout the Semester or by completing an Individual Project.
--> ASSESSMENT THROUGHOUT THE SEMESTER
- Intermediate Test (35%): minimum mark of 7,5;
- Group Project (65%): written report and code (40%) + oral presentation (10%) + individual test (15%).
Conditions associated with assessment throughout the semester:
(i) Maximum number of students who can make up a working group: 5;
(ii) The student must participate in all moments of assessment throughout the semester;
(iii) Group Project: minimum mark of 7,5.
--> INDIVIDUAL EVALUATION
The individual assessment consists of a Test and an Individual Project:
- Test (35%), to be taken on the day of the exam (first season): minimum mark of 7,5
- Individual Project (65%; minimum mark of 7,5):
- Written report and code (40%)
- Oral discussion (25%).
The student must take part in all components of the individual assessment.
2nd SEASON ASSESSMENT
In the 2nd Season, the evaluation is based on a Test and an Individual Project.
--> INDIVIDUAL EVALUATION
- Test (35%), to be taken on the day of the exam (2nd season): minimum mark of 7,5
- Individual Project (65%; minimum mark of 7,5):
- Written report and code (40%)
- Oral discussion (25%).
The student must take part in all components of the individual assessment.
Passing the course: A final mark of 10 or above.
In both forms of assessment, students may be required to take an additional individual evaluation, either oral or written, in order to confirm their mark.
Scale 0-20
- Chopard, B.; Tomassini, M. (2018). An Introduction to Metaheuristics for Optimization, Springer.
- Gutierrez, A. M; Ramirez-Mendoza, R. A.; Flores, E. M.; Ponce-Cruz, P; Espinoza, A.A. O.; Silva, D. C. B. (Eds.) (2020). A Practical Approach to Metaheuristics using LabVIEW and MATLAB (R). Taylor & Francis Ltd.
- Ke-Lin Du; M. N. S. Swamy (2016). Search and Optimization by Metaheuristics: Techniques and Algorithms Inspired by Nature. Birkhäuser.
- Ragsdale, C.T. (2022). Spreadsheet Modeling and Decision Analysis: A Practical Introduction to Business Analytics, 9th Ed. Cemgage Learning.
- Taha, Handy A. (2023). Operations Research: An Introduction, 11th Ed. Pearson. (não tem Goal Programming e tem heurísticas)
- Taha, Handy A. (2017). Operations Research: An Introduction, 10th Ed. Pearson. (tem Goal Programming e tem heurísticas)
- Burke, E. K.; Kendall, G. (Eds.) (2014). Search Methodologies: Introductory Tutorials in Optimization and Decision Support, 2nd edition, Springer.
- Ehrgott, M. (2005). Multicriteria Optimization, 2nd edition, Springer.
- Lobato, F. S.; Valder, S. Jr. (2017). Multi-Objective Optimization Problems: Concepts and Self-Adaptive Parameters with Mathematical and Engineering Applications. Springer Cham.
- Open Access documents such as instructor notes, book chapters, research articles, and tutorials that will be provided via Fenix.
- Siarry, P. (Ed.) (2016). Metaheuristics, Springer.
Unsupervised Learning Methods
LO1: Characterize the main unsupervised data methods
LO2: Use R for unsupervised data analytics
LO3: Evaluate, validate and interpret the results
PC1: Introduction to unsupervised learning methods
PC2: Data reduction techniques (dimensionality)
- Principal components analysis (PCA)
- Data reduction techniques using R
PC3: Clustering techniques
- Hierarchical methods
- Partitioning methods
- Self-organizing maps
- Probabilistic methods
- Quality & Validity of clustering methods
- Clustering techniques using R
PC4: Case studies
Students may choose either Evaluation during the semester or Final Exam.
EVALUATION DURING THE SEMESTER:
- group project with discussion and minimum grade of 8 (50%)
- individual test with minimum grade of 8 (50%)
Approval requires a minimum attendance of 80% of classes, participation in all assessment components, and minimum grade of 10.
EXAM:
The Final Exam is a written exam. Students must achieve a minimum grade of 10 to pass.
In both forms of assessment, students may be called upon to sit an individual examination, either oral or written, to validate their grade.
Nwanganga, F., M. Chapple (2020), Practical Machine Learning in R, 1st Edition, Wiley. Bouveyron, C., G. Celeux, T. B. Murphy, A. E. Raftery (2019), Model-Based Clustering and Classification for Data Science: With Applications in R, 1st Edition, Cambridge University Press. James, G., Witten, D., Hastie, T., Tibshirani, R. (2013), An Introduction to Statistical Learning: with applications in R, New York: Springer. Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E. (2014), Multivariate Data Analysis, 7th Edition, Essex, UK: Pearson Education.
-
Regression Models
LO1. Understand the correlation between variables and simple and multiple linear regression models.
LO2. Know the main parameter estimation methods (OLS and ML).
LO3. Know how to perform regression model assumption analysis, hypothesis testing, and diagnosis.
LO4. Understand the limitations of linear regression using quantile regression and regularization methods.
LO5. Application of the concepts studied; training/test and prediction sets, information and value extraction for real-world data with R software.
S1. Introduction to Regression Models
S1.1. Correlation
S1.2. Simple linear regression
S1.3. Multiple linear regression
S2. Estimation and inference
S2.1 OLS estimation
S2.2 ML estimation
S2.3 Statistical inference
S3. Model diagnosis and validation
S3.1. Assumptions
S3.2 Diagnostic Tests
S3.3. Practical Examples
S4. Extensions of Linear Regression
S4.1 Regularisation Methods (Ridge and Lasso)
S4.2. Quantile Regression
S4.3. Practical Examples
S5. Applications using R for real-world data and scenarios
Assessment throughout the semester includes the following:
a) An individual mid-term test, accounting for 40% of the mark.
b) An individual test, accounting for 60% of the mark, with a minimum mark of 8.5.
Exam assessment: an individual exam covering the entire syllabus, with a minimum mark of 10 (mark rounded to the nearest whole number). The individual test and the exam will be held without the use of handouts, books or other materials; the use of graphing calculators or mobile phones is not permitted; students may only consult the form and tables made available on Moodle for this purpose.
In both cases, students may be called upon to sit an individual oral or written assessment to validate their mark.
Cameron, A.C., Trivedi, P.K. (2008) Microeconometrics: Methods and Applications. Cambridge University Press
Bruce P., Bruce A., and Gedeck P., (2020), Practical Statistics for Data Scientists: 50+ Essential Concepts Using R and Python, 2nd Edition,
O' Reilly Media, Inc. - Robert I. Kabacoff, (2022), R in Action: Data analysis and graphics with R, Third Edition, Manning Publications Co. -
Tom Alby, (2024), Data Science in Practice, CRC Press.
Robert I. Kabacoff, (2022), R in Action: Data analysis and graphics with R, Third Edition, Manning Publications Co. Wooldridge, J.M. (2019), "Introductory Econometrics: A Modern Approach", 7th Ed., Cengage Learning.
Eric Goh Ming Hui, (2019), Learn R for Applied Statistics, Apress.
Daniel J. Denis, (2020), Univariate, Bivariate, and Multivariate Statistics Using R: Quantitative Tools for Data Analysis and Data Science, JohnWiley & Sons, Inc. Ficheiros (slides e scripts) da UC a disponibilizar no Moodle.
Computational Statistics
Learning goals (LG) to be developed :
LG1: Consolidate the use of R software in the RStudio environment
LG2: Know how to calculate probabilities in various contexts, including through simulation
LG3: Be familiar with the most common probabilistic behavior models
LG4: Know how to fit probabilistic models
LG5: Understand the principles of statistical inference
LG6: Know how to choose the most appropriate inferential method for each situation
Syllabus contents (SC):
SC1- Probability theory: definitions, axioms, conditional probability, total probability theorem and Bayes' formula
SC2- Univariate random variables: mass and density functions, distribution function, and parameters. Working with usual random variables. Simulation of RV with a specified distribution.
SC3-Bi and multivariate RVs. Joint probability and distribution functions. correlation and covariation. Independence between RVs. Sample joint distribution.
SC4- Sampling distributions: limit central theorem, theoretical sampling distributions.
SC5- Parameters estimation: point estimation, estimators' properties, maximum likelihood estimators, interval estimation
SC6- Hypothesis testing: types of errors and corresponding probabilities. Test for one and two means. Chi-square of independence. Meaning and computation of p-values.
Students may choose either Assessment Throughout the Semester or Final Exam.
Assessment Throughout the Semester
1. Small assessment components throughout the semester**: 15% of the final grade. These may include quizzes, homework exercises, contributions to the Glossary or Forum on Moodle, etc. About 10 contributions are expected, with the best 8 being counted.
2. Midterm test, weight: 30%, no minimum grade required.
3. Final test, weight: 55%, minimum grade of 8 out of 20 required.
OR
Final Exam, 100%
A student who, during the regular exam period, has not achieved a passing grade but has scores for #1 (ongoing assessment components throughout the semester) and #2 (midterm test), may use the resit exam period (or special exam period, if applicable) as a replacement for the final test. Their final grade in these periods will be calculated according to the rules of assessment throughout the semester. The student must state this intention in writing on the exam they take.
This option is not available for grade improvement purposes.
Speegle, D., & Clair, B. (2021). Probability, Statistics, and Data: A Fresh Approach Using R (1st ed.). Chapman and Hall/CRC. Free access at https://mathstat.slu.edu/~speegled/_book/
Reis, E., Andrade, M., Calapez, T. & Melo, P., Estatística Aplicada, volume 1. 6ª edição. Lisboa. Edições Sílabo., 2015, ISBN 978-972-618-819-3.
Reis, E., Andrade, M., Calapez, T. & Melo, P., Estatística Aplicada volume 2, 6ª edição, Lisboa. Edições Sílabo., 2016, ISBN 978-972-618-986-2.
Verzani, J., Using R for Introductory Statistics, 2nd Edition, Chapman & Hall/CRC, 2014, eBook ISBN 9781315373089, https://cran.r-project.org/doc/contrib/Verzani-SimpleR.pdf
Reis, E., Andrade, M., Calapez, T. & Melo, P., Exercícios de Estatística Aplicada volume 1. 2ª edição, Lisboa. Edições Sílabo., 2012, ISBN 978-972-618-688-5
Reis, E., Andrade, M., Calapez, T. & Melo, P., Exercícios de Estatística Aplicada volume 2. 2ª edição, Lisboa. Edições Sílabo., 2014, ISBN 978-972-618-747-9
Supervised Learning Methods
LG1: Understanding supervised learning methods: scopes of application and procedures
LG2: Use of R software to perform data analysis
LG2: Evaluate and interpret the data analysis results
PC1: Overview of Supervised Learning
Typologies
Learning data
Objective functions
Models' assessment and selection
Notes on statistical inference
PC2: Regression Methods
K-Nearest Neighbor
Regression Trees (using CART algorithm)
PC3: Classification Methods
Naive Bayes
K-Nearest Neighbor
Logistic Regression
Classification Trees (using CART algorithm)
The Course can be assessed using the Assessment throughout Semester or Assessment by Exam.
ASSESSMENT THROUGHOUT SEMESTER:
- group quiz online (40%) with a minimum grade of 9
- individual test (60%) with a minimum grade of 9
Approval requires a minimum grade of 10.
ASSESSMENT BY EXAM:
1st part - individual test (60%)
2nd part -individual practical data analysis test, online, with the R software used in classes (40%)
Students have to achieve a minimum grade of 9 in each part of the exam and a combined minimum grade of 10.
Scale 0-20
Gareth, J., Daniela, W., Trevor, H., & Robert, T. (2013). An introduction to statistical learning: with applications in R. Springer.
Hastie, T., Tibshirani, R., Friedman, J. H., & Friedman, J. H. (2009). The elements of statistical learning: data mining, inference, and prediction (Vol. 2, pp. 1-758). New York: Springer.
Lantz, B. (2023). Machine Learning with R: Learn techniques for building and improving machine learning models, from data preparation to model tuning, evaluation, and working with big data. 4th Edition. Packt Publishing.
Molnar, C. (2020). Interpretable machine learning. Lulu. com.
Field, A., Miles, J., & Field, Z. (2012). Discovering statistics using R (2012). Great Britain: Sage Publications, Ltd, 958.
Larose, D., Larose, C. (2015). Data Mining and Predictive Analytics. John Wiley & Sons.
Network Analysis
By successfully completing this Curricular Unit, students will be able to:
LO1. Classify networks using correlation and clustering coefficients, distances, centrality measures and heterogeneity measures. Evaluate the network robustness.
LO2. Obtain the co-occurrence network associated with a network representing relations. Analyze weighted networks.
LO3. Select random network generation models and characterize random networks.
LO4. Detect communities and evaluate the methods applied to detect communities.
Programmatic Contents (PC):
PC1: Basic Concepts
Elements of a network, subnetworks, density and degree. Bipartite networks.
PC2: Small Worlds
Degree correlation. Paths and distances. Connectivity. Six Degrees of Separation. Clustering coefficients.
PC3: Hubs and Weight Heterogeneity
Centrality measures. Heterogeneity based on degree. Robustness, core decomposition and weight heterogeneity.
PC4: Random Networks
Generation and characterization of random networks.. Watts-Strogatz’s model, Configuration Model, and preferential models.
PC5: Communities
Concepts and related problems. Methods for community detection: bridge removal, modularity optimization, label propagation. Evaluation methods.
Assessment throughout the semester or assessment by exam:
1. Assessment throughout the semester:
a) 5 individual short tests: i) weighting 10%; ii) completed in class throughout the lecturing period; iii) grade for this component: arithmetic mean of the 4 best grades obtained on the 5 short tests;
b) Group project: i) weighting 30%; ii) Groups of 4 students; iii) with oral discussion;
c) Individual final written test: i) weighting 60%; ii) minimum classification of 8.5;
d) Attendance of at least 2/3 of the classes taught;
e) Approval: minimum weighted average of 9.5.
2. Evaluation by Exam (1st and 2nd Season):
a) Individual written exam: i) weight of 100%; ii) approval: minimum classification of 9.5;
In both assessment methods, students may be required to take an individual exam, either oral or written, aimed at validating the grade obtained.
Scale: 0-20 points.
* Menczer, F., Fortunato, S. and Davis, C. A. (2020). A First Course in Network Science, 1st Edition, Cambridge University Press, Cambridge.
* Barabási, A.-L. (2016). Network Science, 1st Edition, Cambridge University Press, Cambridge.
* Newman, M. (2018). Networks, 2nd Edition. Oxford University Press, Oxford.
* Ognyanova, K. (2018). Introduction to R and Network Analysis. SC&I Methods Workshop, Rutgers University.
Fundamentals of Database Management
O1: Develop abstraction mechanisms;
O2: Develop Information Modeling abilities;
O3: Develop the ability to extract data from a database in an efficient way.
P1 - Database Design
P2 Relations and primary keys
P1.2.2 Foreign Keys and Integrity Rules
P1.2.3 Optimizationsand Indexes
P1.2.5 Transctions and Concurrency
P2 S.Q.L
P2. 1 Simpl Querys;
P2.2 Agregate Functions;
P2.3 SubQuerys;
P2.4 Triggers and Stored Procedures;
Assessment is done through exam, sseason 1, season 2 and special season.
Bibliography-Ramos, P, Desenhar Bases de Dados com UML, Conceitos e Exercícios Resolvidos, Editora Sílabo, 2ª Edição, 2007
-Perreira, J. Tecnologia de Base de Dados" FCA Editora de Informática, 1998
-Damas, L. SQL - Structured Query Language " FCA Editora de Informática, 2005 (II)
http://plsql-tutorial.com/.
-Date, C.J. "An introduction to Database Systems" Addison-Wesley Publishing Company, sexta edição, 1995 (I.2, I.3, I.4, II);
-Booch, G., Rumbaugh, J., Jacobson, I "The Unified Modeling Language User Guide" Addison-Wesley Publishing Company, 1999 (I.1);
-Nunes, O´Neill, Fundamentos de UML, FCA, 2002
Web Interfaces for Data Management
After finishing this unit a student should be able to:
LG1. Know and understand basic concepts and technologies for Web development.
LG2. Know and understand interface technologies between a Web application and a Database.
LG3. Model and develop a Web application allowing to manage persistant data from human interaction with software on the Web.
CP1 [Introduction]
- The history of the Web;
- Previous and actual programming languages for the web;
- W3C standards;
CP2 [Modelling and programming a Web application]
- Client-server architecture;
- MVC architecture for the Web.
- Main graphical formatting languages for the Web;
- Libraries for graphical formatting;
- Main programming languages for the Web;
- Libraries for programming for the Web;
- Introduction to security on the client and on the server side.
CP3 [Database access]
- Database access from the Web;
- Data model on the Web application and corresponding interaction with the Database.
CP4 [Data Storage and Management]
- Storage of Web data in a Database;
- Data management.
Given the practical nature of the content taught, assessment will take place throughout the semester, and there will be no exam-based assessment.
First assessment period:
Exercises (10%), completed in class and in groups.
Project (40%), completed in groups, including a report and software. Mandatory, with a minimum grade of 7 out of 20. If there are doubts regarding the project assessment, the teaching team may call the group for an oral discussion.
Test (50%), completed individually. Mandatory, with a minimum grade of 7 out of 20. If the Test grade is more than 5 points lower than the Project grade, the Project grade will be reduced so that the difference between the two grades is at most 5 points.
Second assessment period:
Possibility of retaking the Test (50%) for those who failed the Test taken in the first assessment period.
The grades obtained throughout the semester for the Exercises (10%) and the Project (40%) will be taken into account. Class attendance is not mandatory.
Special assessment period:
Students registered for the special assessment period will complete only the Test (50%) and Project (50%) components.
Improvement of the overall assessment grade may only be carried out in the following academic year.
Mitchell, R. (2016). Web Scraping with Python: Collecting Data from the Modern Web. Ed. O?Reilly Media, Inc. ISBN-13: 978-1491910290. ISBN-10: 1491910291.
Vincent W. S. (2018). Build websites with Python and Django. Ed: Independently published. ISBN-10: 1983172669. ISBN-13: 978-1983172663.
Dean J. (2018). Web Programming with HTML5, CSS, and JavaScript. Ed: Jones & Bartlett Learning. ISBN-13: 978-1284091793. ISBN-10: 1284091791.
Ryan J. (2013). A History of the Internet and the Digital Future. Ed: Reaktion Books. ISBN-13: 978-1780231129
Lambert M. and Jobsen B. (2017). Complete Bootstrap: Responsive Web Development with Bootstrap 4. Ed: Impackt Publishing. ISBN-10: 1788833406. ISBN-13: 978-1788833400.
Downey A. B. (2015). Think Python: How to Think Like a Computer Scientist. Ed: O'Reilly Media. ISBN-10: 1491939362. ISBN-13: 978-1491939369.
Stocastic Modelling
On completion of this course, students should:
LG1. Understand the principles and methods of stochastic simulation;
LG2. Be able to develop efficient algorithms for generating pseudorandom numbers;
LG3. Be able to apply the Monte Carlo method;
LG4. Understand and be able to apply different Monte Carlo via Markov Chains methods;
LG5. Be able to implement resampling techniques;
LG6. Be able to simulate a real system through discrete event simulation;
LG7. Be able to analyze and evaluate simulation results;
LG8. Be able to implement efficient stochastic simulation algorithms in R.
S1. Introduction to Simulation in Data Science
S2. Generation of Pseudo-Random Numbers
- Linear Congruential Method; Inverse Transformation Method; Acceptance/Rejection Method; Other Transformations; Mixtures
S3. Monte Carlo Methods in Statistical Inference
S4. Markov Chain Monte Carlo (MCMC) Methods
S5. Resampling Methods
- Bootstrap; Cross-Validation
S6. Discrete Event Simulation
- Assessment throughout the semester
Assessment throughtout the semester requires attendance at a minimum of two-thirds of the scheduled classes and comprises the following components:
• Delivery of two classes by students, working in groups, on topics assigned by the lecturer – 40%
• Class participation and contribution – 10%
• Individual written test – 50%
To pass the course, students must obtain a minimum mark of 8.5/20 in the written test and an overall final mark of at least 10/20.
- Assessment by Final Examination
Students opting for assessment by final examination will be assessed on the basis of:
• Written examination – 50%
• Individual essay – 50%
The essay should demonstrate critical analysis, a sound understanding of the course content, and the appropriate application of the methods and techniques covered in the course.
- Supplementary Oral Examination
The Course Unit Coordinator may require a supplementary oral examination in relation to any assessment component specified in this Course Unit Outline whenever this is deemed necessary to confirm the acquisition of the knowledge and competences being assessed or to clarify discrepancies between different assessment components.
The supplementary oral examination does not constitute a separate assessment component, carries no independent weighting in the final course mark, and serves exclusively to validate the mark awarded for the relevant assessment component. The mark previously awarded to that component shall be regarded as provisional until the supplementary oral examination has taken place and may be confirmed or revised in accordance with the outcome of the validation process and the applicable assessment criteria.
Students selected for a supplementary oral examination will be notified at least 24 hours in advance, with the date and time of the examination specified. Failure to attend without an acceptable justification will prevent validation of the relevant assessment component, and the consequences applicable to non-completion or withdrawal from that assessment component shall apply.
Templ, M. (2016). Simulation for Data Science with R. Packt Publishing Ltd:Birmingham, Uk.
Rizzo, M. L. (2008). Statistical Computing with R. Chapman & Hall/CRC.
Robert, C. P. and Casella, G. (2010). Introducing Monte Carlo Methods with R. Springer-Verlag.
Wickham, H. and Grolemund , G. (2017). R for Data Science. O'Reilly Media Inc.
Symbolic Artificial Intelligence for Data Science
The course introduces the major themes of (mostly) Symbolic Artificial Intelligence and Machine Learning, from an essentially applied perspective, bearing in mind the major context provided by the data science degree, the knowledge and skills acquired in the other courses, and the fundamental objectives and requirements of the data science degree.
The three major topics of the program are logic programming, mostly symbolic adaptive techniques for the representation of adaptive world models, and symbolic machine learning algorithms to learn world models.
After the students have completed the course, they must
? Be fully aware of the existence of mainly symbolic paradigms for the representation and autonomously learning of adaptive world models.
? Have mastered the capability to decide whether to use the paradigms learned in the course to application problems / domains whenever suited.
Overview of the Curricular Unit: the need, advantages and disadvantages of essentially symbolic technologies for representing and learning adaptive models of reality, and the role of each programme component in the desiderata of the chair.
Programming in logic to represent models of reality and to reason with them.
Representation and reasoning based on fuzzy sets and in fuzzy logic to represent essentially symbolic adaptive models and reason with them.
Representation and reasoning based on cases to represent essentially symbolic adaptive models and reason with them.
Introduction to Explainable AI and its characteristics and application domains.
Concepts of Responsible AI.
In semester assessment, students will have to take:
- Individual written test on the entire CU programme (60%) - occurring during exams' period (1st or 2nd exam).
- (Group) research work on one of the CU topics, with a report and an oral presentation (40%). The oral presentation is done in class time during the semester. The grade of the research work is split 50% for each item and the members of the group may have different grades.
Both assessment components on semester evaluations have a minimum mark of 8.
Alternatively, students can take only one exam (100%), which can be at both dates of exams.
At special exams' period the students take the exam (100%).
Whenever there are doubts regarding the authorship or mastery of the knowledge demonstrated in an assessment component, the student may be required to take an individual oral or written exam to validate the corresponding grade.
Logic Programming and Inductive Logic Programming:
Ivan Bratko. 2011. Prolog Programming for Artificial Intelligence (4th Edition). Pearson Education Canada (International Computer Science Series).
Fuzzy Systems:
Guanrong Chen, and Trung Tat Pham. 2005. Introduction to Fuzzy Systems. CRC Press.
Case based reasoning:
Michael M. Richter, and Rosina Weber. 2013. Case-Based Reasoning. A Textbook. Springer-Verlag Berlin Heidelberg
Lynne Billard, Edwin Diday. 2007. Symbolic Data Analysis: Conceptual Statistics and Data Mining, John Wiley & Sons, Ltd, Chichester, UK
Applied Project in Data Science
LO1: Apply methodologies appropriate to the challenge at hand
LO2: Develop skills in cleaning, preprocessing, and integrating real data
LO3: Transform, summarize and visualize data effectively
LO4: Analyze data and interpret results critically
LO5: Communicate results clearly and in a structured manner
CP1: Preliminary data analysis. Exploring libraries for manipulating, cleaning, visualizing, and processing real data, with a focus on preparing data for analysis.
CP2: Variable Analysis and Feature Engineering. Identifying key variables and applying effective selection and transformation techniques to enhance model accuracy and the overall quality of analysis.
CP3: Visualization of univariate and multivariate data.
CP4: Analysis, interpretation, and effective communication of results to support data-driven decision-making.
CP5: Development of an applied project using real-world data, encompassing all stages of the analytical process — from problem definition to the presentation and defense of data-driven solutions
Taking into account the practical nature of the course, assessment is conducted continuously throughout the semester and does not include a final exam. Evaluation is based on two main components:
a) Group quizzes with consultation (40%)
a.1) Two online group quizzes, carried out in class (5% each);
a.2) One comprehensive online group quiz, on the full course content (30%)
b) Group Project (60%)
b.1) Ongoing weekly progress and participation (10%)
b.2) Final written report and oral presentation (50%)
Minimum Requirement: Students must achieve a minimum grade of 9 (out of 20) in each component to pass the course.
(1) Hastie, T., Tibshirani, R., Friedman, J. H., & Friedman, J. H. (2009). The elements of statistical learning: data mining, inference, and prediction (Vol. 2, pp. 1-758). New York: Springer.
https://link.springer.com/book/10.1007/978-0-387-84858-7
(2) James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning: with applications in R (Vol. 103). New York: Springer.
https://link.springer.com/book/10.1007/978-1-0716-1418-1
(3) Cady, F. (2024). The data science handbook. John Wiley & Sons.
https://www.wiley.com/en-us/The+Data+Science+Handbook%2C+2nd+Edition-p-9781394234493
(4) Boehmke, B., & Greenwell, B. M. (2019). Hands-on machine learning with R. Chapman and Hall/CRC.
https://bradleyboehmke.github.io/HOML/, https://bradleyboehmke.github.io/HOML/
(5) Wes McKinney (2022), Python for Data Analysis, 3rd Edition, O'Reilly Media, Inc. (https://wesmckinney.com/book/, https://github.com/wesm/pydata-book)
(1) Harrison, M., & Petrou, T. (2020). Pandas 1. x Cookbook. Packt Publishing.
(2) Mukhiya, S. K., & Ahmed, U. (2020). Hands-On Exploratory Data Analysis with Python: Perform EDA techniques to understand, summarize, and investigate your data. Packt Publishing Ltd.
https://github.com/PacktPublishing/hands-on-exploratory-data-analysis-with-python
(3) Gagolewski, M. (2022). Minimalist Data Wrangling with Python. arXiv preprint arXiv:2211.04630.
https://datawranglingpy.gagolewski.com/
(4) Vasconcelos, J. B., & Barão, A. (2017). Ciência dos dados nas organizações. FCA Editora.
Applied Final Project in Data Science
At the end of the course, each student should be able to:
LO1. Define the objectives and formulate the CD tasks that allow the client to extract the desired knowledge.
LO2. Define the data variables and metadata that lead to the required knowledge.
LO3. Plan the different phases of project development.
LO4. Process the data with the most appropriate Data Science tools to achieve the proposed objectives.
LO5. Produce data visualisations and documents suitable for correctly communicating the results obtained.
LO6. Solve problems inherent in using real data from an ethics-by-design perspective.
The programme contents (CP) are as follows:
CP1. Introduction to the proposed challenges (projects) and organisation of project teams.
CP2. Information research methodologies for framing the project theme.
CP3. Practical approaches to project development from an ethics-by-design perspective.
CP4. Tools for each stage of project development.
CP5. Usual models for communicating data and results.
Being a project-based course, there is no 100% examination. Assessment runs throughout the semester and consists of the different stages of project development, where:
(i) Each stage is marked by a deliverable (written or presented in class) (E) with feedback.
(ii) An intermediate presentation (A1) with feedback.
(iii) A final presentation (A2) with discussion.
(iv) A poster (informative) (P).
(v) A final project report (R).
The grade will be the result of E x 0.15 + A1 x 0.10 + P x 0.10 + A2 x 0.30 + R x 0.35.
Dependente dos temas específicos do desafio em que cada grupo de estudantes irá desenvolver os trabalhos do projeto.
Voeneky, S., Kellmeyer, P., Mueller, O., & Burgard, W. (Eds.). 2022. The Cambridge Handbook of Responsible Artificial Intelligence: Interdisciplinary Perspectives. Cambridge: Cambridge University Press.
Witten, I. H., Frank, E., Hall, M. A., & Pal, C. J. 2016. Data Mining: Practical machine learning tools and techniques. Morgan Kaufmann.
Provost, F., & Fawcett, T. 2013. Data Science for Business: What you need to know about data mining and data-analytic thinking. O'Reilly Media, Inc.
Introduction to Deep Learning
LO1: Explain the fundamental principles of deep learning and its evolution within the field of artificial intelligence.
LO2: Formulate deep learning problems by identifying data, models, parameters, loss functions, evaluation metrics, and generalization criteria.
LO3: Describe and analyze dense neural networks, including activation functions, depth, function composition, representational capacity, expressiveness, and practical limitations.
LO4: Train neural networks using optimization methods, understanding the role of backpropagation, weight initialization, and training stability.
LO5: Evaluate and improve the performance of deep learning models, identifying limitations, risks of overfitting and underfitting, and trade-offs between performance, interpretability, and computational cost.
LO6: Understand modern architectures, including convolutional neural networks, residual networks, Transformers, pre-trained models, and generative models such as VAEs and diffusion models.
P1 — Deep Learning Foundations: machine learning context; learning paradigms; data, models, parameters, loss functions, and generalization.
P2 — Neural Networks: shallow and deep networks; activation functions; depth, representation, and expressiveness.
P3 — Network Training: loss functions; gradient-based optimization; backpropagation; and weight initialization.
P4 — Evaluation and Regularization: training, validation, and testing; metrics; overfitting, underfitting, weight decay, dropout, data augmentation, and early stopping.
P5 — Convolutional Architectures: convolution, filters, pooling, CNNs, residual networks, transfer learning, and encoder–decoder models.
P6 — Transformers: tokenization, embeddings, attention, Transformers, fine-tuning, inference, and large language models.
P7 — Generative Models: autoencoders, latent spaces, VAEs, reconstruction, sampling, and diffusion models.
The course may be completed through one of two assessment methods:
1 – Continuous Assessment
Continuous assessment consists of:
- (80%) Two tests: one midterm test, held during the semester on a date to be announced, and a second test held on the first examination date;
- (20%) Online quizzes based on the practical assignments, completed through the e-learning platform. Attendance at the corresponding practical session is mandatory for the assessment to be considered.
The final grade is calculated as a weighted average of these two components according to the percentages above, subject to the following conditions:
- Students must obtain a minimum grade of 7.5/20 in both tests to pass the test component;
- Absence from three or more practical sessions results in automatic failure of the practical assessment component;
- For the purpose of the final grade calculation, practical assignments corresponding to missed sessions are awarded a grade of zero.
Failure in either component of the continuous assessment requires the student to take the final examination assessment.
2 – Final Examination Assessment
The final examination assessment consists of a single exam taken during any of the available examination periods and accounts for 100% of the final grade.
At any time, and regardless of the assessment method, students may be required to undertake an oral examination to confirm the awarded grade, verify the authorship of submitted work, and/or demonstrate the competencies assessed in the course.
- Prince, S. J. D. (2023). Understanding Deep Learning. MIT Press. (acesso aberto: https://udlbook.github.io/udlbook/)
- Zhang, A., Lipton, Z. C., Li, M., & Smola, A. J. (2023). Dive into Deep Learning. (acesso aberto: https://d2l.ai)
- Raschka, S., Liu, Y., & Mirjalili, V. (2022). Machine Learning with PyTorch and Scikit-Learn. Packt.
- Tunstall, L., von Werra, L., & Wolf, T. (2022). Natural Language Processing with Transformers. O'Reilly.
- Goodfellow, I., Bengio, Y., & Courville, A. (2016). *Deep Learning*. MIT Press.
- Documentação oficial do PyTorch e tutoriais da Hugging Face.
Objectives
The Bachelor's Degree in Data Science aims to train professionals capable of handling the growing amount of data generated by modern society. This degree aims at acquiring analytical skills and applying appropriate methodologies to the data cycle.
In this course, skills are developed in Computer Science and Programming Technologies, for the development of basic software tools for Data Science; Information Systems, for designing and managing Databases, storing and processing large volumes of data (Big Data); Artificial Intelligence; Statistics and Data Analysis, to analyze and classify data, identify patterns, make forecasts and simulations. Additionally, students are promoted and made aware of the importance of Security, Ethics, and Privacy.
Many course units include group courseworks, which require and promote teamwork. In addition to theoretical-practical classes, many course units have practical-laboratory classes, in which the methodologies and technologies relevant to Data Science are applied.
The degree also includes applied project course units, in which student groups apply the studied contents to real data and problems. In the “Final Applied Project in Data Science”, each group of students develops a project proposed by an external entity.
The bachelor should be able to attain the learning outcomes:
Skills:
- be able to collect, clean, transform, an query data;
- be able to organize, summarise, visualize data and outcomes;
- be able to select and apply the appropriate methodologies to perform data analysis, statistical inference, and predictive and prescriptive analysis;
- be able to implement algorithms in a general purpose language;
- be able to evaluate and reflect on the level of security, data protection and privacy of a specific technological solutions.
Competencies:
- be able to develop data-driven analysis;
- be able to search and evaluate scientific knowledge;
- be able to work within multidisciplinary teams, while communicating results to stakeholders.