The course introduces the core methodologies behind automatic approaches aimed at making sense of social and psychological aspects of human behaviour. In particular, the course shows 1) how to design and organise the observation of human behaviour in view of the application of automatic approaches, 2) how to apply psychometric instruments for the quantitative analysis of social and psychological phenomena, and 3) how to apply basic statistical techniques to human behaviour analysis and understanding. The course is interdisciplinary and it requires the acquisition of both computing and social psychological notions. The application areas to which the course is relevant include, e.g., social robotics, user experience analysis, social media analytics, surveillance and e-health (the list is not exhaustive).

Three hours per week.

Computational Social Intelligence (M)

None

Examination 80%, Report 20%.

The aim of the course is to introduce the students to the main computational methodologies for automatic analysis of human behaviour. In particular the course teaches how to design and organise the observation of human behaviour in view of the application of computational approaches. Furthermore, it shows how to quantify social and psychological phenomena through the application of standard psychometric questionnaires. Finally, it introduces basic methodologies - based on machine learning and statistics - aimed at mapping behavioural observations into high-level interpretations of human behaviour that take into account social and psychological aspects of human-human and human-machine interactions.

By the end of this course, the students will be able to:

1. Design and organise the collection of behavioural data in view of the application of statistical and computational methodologies for human behaviour understanding;

2. Measure social and psychological constructs - in quantitative terms - through the adoption of standard psychometric questionnaires; 

3. Apply basic statistical methodologies (e.g., k-Means and Naïve Bayes Classifier) to automatically map behavioural observations into social and psychological constructs.