Location-based information services for mobile users are becoming widely available. These services, however, only take the user's position into consideration. Therefore, it is often necessary for users to execute bothersome operations to obtain the information they need in content-heavy areas. To address this problem, the authors have developed TPOCAST, a TPO-based personalized information delivery platform for mobile users. It provides personalized information to the mobile users on the basis of current time, geographic positions, and individual preferences depending on the occasion. The key to this platform is the automatic learning capability of user preferences by monitoring user actions. An experimental service for tourists using TPOCAST was performed in Matsue City (Japan). This paper reports details on the experiment.
Abstract