User profile modelling based on mobile phone sensing and call logs

Abstract

There are remaining questions concerning user profile modelling in the mobile advertising domain. The research question addressed in this paper is how to design a specific user profile model, that is a simplified model in terms of the amount of user data to be collected, that considers relevant aspects of mobile advertising such as social and personal context, and user privacy preservation. To address this question, a new user profile model consisting of three phases was proposed: (1) data collection, (2) integration and normalization of collected data, and (3) inference of knowledge about the mobile user’s profile. The most significant contributions of the proposed model are a simplified user profile model approach which tackles the dependency on other data sources like OSN platforms and local data gathering and storage that contributes to the user privacy-preserving since the user can exert more control over his/her personal data.