Abstract

Many location-based software applications have been developed for mobile devices. Consequently, location-based service providers often have a detailed trajectory history of their service recipients. The collected spatial-temporal information of their service recipients can be invaluable for other organizations and companies in many ways; for example, it can be used for direct marking, market analysis, and consumer behaviour analysis. Yet, releasing the spatial-temporal data together with other user-specific data in its raw format often leads to privacy threats to the service recipients. In this thesis, we study the problem of spatial-temporal data publishing with the consideration of preserving both privacy protection and information utility for data mining. The contributions are in twofold. First, we propose a service-oriented architecture to determine an appropriate location-based service provider for a given data request. Second, we present a border-based data anonymization method to transform a raw spatial-temporal data table into an anonymous version that preserves both privacy and information utility. Experimental results suggest that our proposed method can efficiently and effectively preserve the information required for data mining.