Ai Methods for Sensor-Based Activity Recognition and Context-Awareness

The goal of this course is to introduce the main concepts about techniques that analyze data from several sensors (e.g., the ones on mobile/wearable devices, or in smart-homes environments) to automatically detect the user's context (e.g., the activities performed, the visited places, the surrounding environment). These methods are crucial to develop applications that automatically adapt to the users's context. In particular, this course will focus on one of the most interesting problems in this area: the automatic recognition of human activities. Indeed, monitoring the activities that humans perform in ther daily life enables several important applications, that range from healthcare (e.g., remote monitoring of elderly subjects) to well-being (e.g., monitoring the physical activity level). Activity recognition techniques may be based on machine learning, knowledge-based reasoning, and hybrid approaches (i.e., a combination of machine learning and knowledge-based reasoning). This course will present the main methods and research results in this area, and it will discuss the most relevant open research problems.