|
Abstract Travel path prediction is to predict the forthcoming locations of the human travel paths, which usually follow certain patterns; e.g., a path leads to a landmark such as a park or mall, follows a highway or street, or brings to an event like a ball game or concert. Path prediction is useful and can be applied to several applications such as traffic or city planning, mobile advertisements, and travel guidance & recommendations. This research tries to predict human travel paths based on traffic flows because the volume of a traffic flow usually represents the popularity of that traffic flow. Other than the concern of prediction accuracy, privacy and security are also important concerns of human travel path prediction. Instead of sending the user paths to the servers, which has the risk of potential privacy and security infringement, the server sends the 3D path matrices to the users and the path prediction is performed at the user side. Therefore, user privacy is strictly enforced and security is better protected by using our method. Keywords Global positioning system (GPS), handheld/mobile/smartphone computing, location-based services (LBS), path prediction, and human behavior recognition References
|