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.