Capturing and understanding users' context is key to the success of applications such as computational advertising and recommender systems. Currently, context is usually inferred from a user's interaction history with Web content. As an ever-increasing portion of Web interaction occurs in a mobile context, the physical and social environment in which the interaction occurs is a key factor of user context and holds additional signals of user intent. In this talk, I will discuss recent work on understanding user behavior based on recorded traces of the cyber, social and physical component of users interaction with the in indoor environment (i.e., shopping malls), based on an extensive dataset of Wi-Fi logs. We will draw lessons for the future of indoor recommender services and context-aware mobile applications.
Paper
Full text
Understanding Indoor Behavior: Where, What, with Whom?
Semantic Scholar · Computer Science · 2017
Abstract
Capturing and understanding users' context is key to the success of applications such as computational advertising and recommender systems. Currently, context is usually inferred from a user's interaction history with Web content. As an ever-increasing portion of Web interaction occurs in a mobile context, the physical and social environment in which the interaction occurs is a key factor of user context and holds additional signals of user intent. In this talk, I will discuss recent work on understanding user behavior based on recorded traces of the cyber, social and physical component of users interaction with the in indoor environment (i.e., shopping malls), based on an extensive dataset of Wi-Fi logs. We will draw lessons for the future of indoor recommender services and context-aware mobile applications.