Topics in Intelligent Systems

Code:

M.IA002

Acronym:

TSI

Keywords
Classification	Keyword
OFICIAL	Informatics Engineering
OFICIAL	Computer Science

Instance: 2025/2026 - 1S

Active?	Yes
Responsible unit:	Department of Informatics Engineering
Course/CS Responsible:	Master in Artificial Intelligence

Cycles of Study/Courses

Acronym	No. of Students	Study Plan	Curricular Years	Credits UCN	Credits ECTS	Contact hours	Total Time
M.IA	63	Syllabus	1	-	6	42	162

Teaching language

English

Objectives

This course covers some topics on intelligent systems, emphasizing classical methods from decision theory and statistical learning; but also methods for knowledge representation, ontologies, and semantic models. It is intended to simultaneously study classical methods and more recent techniques while fostering critical reading of scientific articles for the reproduction of relevant methods in the field.

Learning outcomes and competences

The course focuses on statistical learning and methods for knowledge representation. At the end of the course, students should be capable of:

• Critically analyze decision rules or models.
• Develop and implement intelligent systems that involve uncertainty or probabilistic data.
• Apply models for complex representations in decision-making or classification.
• Evaluate and assess the performance of intelligent systems.

Working method

Presencial

Pre-requirements (prior knowledge) and co-requirements (common knowledge)

Linear Algebra, Statistics and Probabilities

Program

Part 1:
“Bayes decision theory” - decision models that deal with uncertainty.
“Gaussian classifier” - optimal models for binary classification.
“Maximum likelihood estimation” - estimation models based on observed data.
“Expectation-maximization (EM)” - classical iterative method for estimation of models that maximize observed data likelihood.

Part 2:
“Representation of Knowledge” - symbols, relations, and logic as a way to capture domain knowledge (vs. statistical data).
“Ontologies and Semantic Models” - classes, properties, and constraints; enable shared understanding and reasoning.
“Reasoning and Querying” - inference with description logics; querying structured knowledge using SPARQL.
“Applications and Integration” - using ontologies for explainability, interoperability, and as a bridge to combine symbolic knowledge with probabilistic models (e.g., BN + KG).

Mandatory literature

Richard O. Duda, Peter E. Hart, David G. Stork; Pattern Classification, Wiley-Interscience Publication, John Willey & Sons, Inc., 2nd edition, 2001
Christopher M. Bishop; Pattern Recognition and Machine Learning, Springer-Verlag New York Inc., 2006

Comments from the literature

Additional bibliography presented during classes.

Teaching methods and learning activities

The course combines theoretical teaching of fundamental concepts with a practical component. In the “hands-on” component there will be exercises to be explored in the classroom, and others as (individual) homework assignments, that constitute a part of the final course evaluation.

The principal component of the course evaluation is a (group) project where students can apply knowledge acquired during classes with other relevant concepts. The “Project-Based Learning” (PBL) approach fosters students' autonomy and independence in acquiring knowledge, facing real challenges, and promoting active learning. Reading of relevant literature is encouraged to develop a critical perspective on research topics of intelligent systems; a key skill for a successful research career.

Evaluation Type

Distributed evaluation without final exam

Assessment Components

Designation	Weight (%)
Trabalho prático ou de projeto	60,00
Trabalho escrito	40,00
Total:	100,00

Amount of time allocated to each course unit

Designation	Time (hours)
Elaboração de projeto	72,00
Frequência das aulas	36,00
Estudo autónomo	48,00
Total:	156,00

Eligibility for exams

An enrolled student obtains attendance if he/she has a minimum grade of 7 on both evaluation components.

Calculation formula of final grade

Final Grade = 40% * Homework assignments + 60% * Project

Special assessment (TE, DA, ...)

Working students and equivalent status holders, who are exempt from regular classes, are not however exempt from regular homework assignments. They can, if they wish and as agreed with the instructors, present progress of their work periodically. The flexibility also extends to the schedules for final project presentations, allowing these students to present their projects at alternative times, as agreed with the instructors. This approach ensures that all students, regardless of their special status, have the same opportunities to demonstrate the progress and results of their work.

Classification improvement

Project evaluations cannot be improved.