Adaptive Interactive Systems

This course is about the design and evaluation of interactive systems that automatically adapt to users and their context. It discusses the layered design and evaluation of such systems. It shows how to build models of users, groups and context, and which characteristics may be useful to model (including for example preferences, ability, personality, affect, inter-personal relationships). It shows how adaptation algorithms can be inspired by user studies. It covers standard recommender system techniques such as content-based and collaborative filtering, as well as research topics such as person-to-person recommendation, task-to-person recommendation, and group recommendation. It also discusses explanations for adaptive interactive systems and usability issues (such as transparency, scrutability, trust, effectiveness, efficiency, satisfaction, diversity, serendipity, privacy and ethics). The course content is presented in the context of various application domains, such as personalized behaviour change interventions, personalized news, and personalized e-commerce.

===

Course Goals

1. Analyse and explain the need for adaptation
2. Choose user, context, group characteristics to model
3. Decide how to model these, critically analysing the (dis)advantages of using different techniques
4. Design adaptation mechanisms, critically analysing the (dis)advantages of using different techniques
5. Explain and show the user experience of an adaptive system
6. Model the implementation of an adaptive system
7. Consider quality aspects of an adaptive system and decide which to focus on and how to measure those
8. Evaluate an adaptive system using layered evaluation
9. Reflect on privacy and ethical issues