Computational Model of Human Routine Behaviours
Nikola Banović, Jennifer C. Mankoff, Anind K. Dey · Oxford University Press eBooks · 2018
Computational Interaction enables a future in which user interfaces (UI) learn about people’s behaviours by observing them and interacting with them to help people to be productive, comfortable, healthy, and safe. However, this requires technology that can accurately model people’s behaviours. This chapter focuses on human routine behaviours enacted by people as sequences of actions performed in specific situations, i.e. behaviour instances, and presents a probabilistic, generative model of human routine behaviours that can describe, reason about, and act in response to people’s behaviours. We holistically define human routine behaviours to constrain the patterns extracted from the data, match routine behaviours, and estimate the likelihood that people will perform certain actions (in different situations) in a way that matches their demonstrated preference. The chapter illustrates how computational models of routines support stakeholders in making sense of stored logs of human behaviour, and designing UIs that respond to those behaviours.