The Emotion Identification through Biological Signals (EIBS) module introduces students to advanced methods in affective computing for the analysis and recognition of human emotions from biological signals. The course covers state-of-the-art approaches for feature extraction from physiological and behavioural data, with a focus on tools and practical techniques to link biological responses to emotional states. Designed for students with prior knowledge in artificial intelligence and basic programming, the module aims to provide hands-on skills for research and the development of intelligent systems sensitive to human behaviour.
Target audience
Postgraduate students with previous coursework on AI (basics) and programming skills
Prerequisites
Basic programming skills and AI (basic courses)
Intended learning outcomes
Students will be able to apply state-of-the-art tools to extract and analyse features of human behaviour.
Teaching and learning methods
The sessions will consist of asynchronous learning content
Assessment
Multiple choice quizzes at the end of each session
Syllabus
Introduction to affective computing
- Definitions, application domains and open challenges
Biological signals for emotion analysis
- Type of signals, characteristics and limitations of different measures
Signal processing
- Filtering, normalisation and noise handling
Feature extraction
- Techniques for extracting features from physiological and behavioural data
- Overview of available open-source tools
Machine Learning methods for emotion identification
- Traditional approaches and deep learning
- Evaluation metrics for models
Use cases and future perspectives
- Practical applications (human-computer interaction, healthcare, gaming, etc.)
- Ethical and societal implications
Instructors
Fabio Salice & Sara Comai
Politecnico di Milano
About Instructor
