Nowadays, Mobile phones have different abilities like satellite positioning which allows calculating the users’ positions. These positions can be used for serving different location-based services for users. Therefore, these services should consider users’ interests and characteristics. This information is usually given manually by the users. If this information is extracted users’ behavior, these services can get interests information automatically. In addition, mobile devices can log calculated users’ locations. Then, each user’s trajectories can be achieved by connecting these locations considering temporal order. Information like interested locations, moving behavior and selected path by users can be extracted using these trajectories. This information can be used for applications like predicting users’ future location. Two methods for extracting interested locations for users are developed in this paper which can be a measure for the users’ behavior. These methods are evaluated using data for Beijing. First method extracts users’ interested locations without considering the time that the users were, but second method considers the presence time for the logged positions for extracting interested locations. Results show that second method has better functionality, because this method considers the trajectory data completely and number of extracted interested locations for second method was more than first method. These algorithms can be used for improving location-based services.