Inferring Long-term User Properties Based on Users' Location History

Recent development of location technologies enables us to obtain the location history of users. This paper proposes a new method to infer users’ longterm properties from their respective location histories. Counting the instances of sensor detection for every user, we can obtain a sensor-user matrix. After generating features from the matrix, a machine learning approach is taken to automatically classify users into different categories for each user property. Inspired by information retrieval research, the problem to infer user properties is reduced to a text categorization problem. We compare weightings of several features and also propose sensor weighting. Our algorithms are evaluated using experimental location data in an office environment.

• Type: Conference
• Year: 2007
• Where: IJCAI
• Authors: Yutaka Matsuo, Naoaki Okazaki, Kiyoshi Izumi, Yoshiyuki Nakamura, Takuichi Nishimura, Kôiti Hasida, Hideyuki Nakashima
• Keywords: Artificial Intelligence, IJCAI 2007, Infer, Respective Location Histories, Text Categorization Problem