Mining User Daily Behavior Based on Location History

With the development of mobile internet and increasing popularity of location aware smart phones, we are enabled to log users' location histories, which are the basis of a variety of location-based services. In this paper, we aim to mine user daily behavior based on a user's location history. As we know, there are regularities in people's daily activities, especially people's daily travel experience. Such regularity is significant to service providers, by recommending potential friends or other information with high relevance to users. Therefore, an approach, namely time- clustering-based behavior analysis (TCBA) is proposed to model each individual's location history and mine the regularity in daily activities. By this approach, we can solve the following queries in a user's daily life: 1) Given a specified time, such as dinner time or working time, what places a user often goes to? 2) What's the regularity of a user's daily life? By using this approach, we can recommend users a convenient route to company in advance, or friends with the same regularity.