轨迹数据挖掘是提高位置感知计算智能化程度和交互效率的有效途径。然而，现有轨迹数据挖掘工作存在两方面问题：第一，现有工作主要基于精确轨迹数据（如GPS轨迹数据），缺乏面向高不确定性的蜂窝基站轨迹数据的挖掘算法研究；第二，现有工作主要针对运动规律，未挖掘出轨迹数据中隐含的大量深层次用户语义信息（如目的意图、生活习惯、社会关系等），导致难以准确估计用户状态和推断用户需求。针对以上问题，本项目提出基于蜂窝基站轨迹数据挖掘的语义化位置感知计算方法。研究适应蜂窝轨迹数据高不确定性的数据抽象方法和数据挖掘算法，从目的、行为、关系三方面对基于轨迹数据挖掘的用户语义推理技术进行研究，探索用户语义对信息适应的影响机制以及面向语义化位置感知应用的交互设计方法。为实现普适计算环境下智能、自然和高效的位置感知计算，促进其在电子服务、智能交通、环境智能、数字娱乐等领域中的应用提供理论依据和技术基础。

Trajectory data mining is regarded as an effective approach to improve the intelligence level and interaction efficiency for location-aware computing. However, there are two problems with the existing works. First, most of the existing works are based on precise trajectory data (e.g. GPS trajectory data), and there lacks the research about the algorithms for mining highly imprecise cellular trajectory data. Second, the existing works mostly focus on mobility regularity mining, and have not extracted the implicit personal semantic information (e.g. intention, living habits, social relations, etc.). As a result, it is very difficult to accurately estimate users' condition and infer users' demand. Aiming at these problems, this project proposes the semantic location-aware computing based on cellular trajectory data mining. This project will research on the data abstraction approaches and data mining algorithms which can adapt to the high uncertainty of the cellular trajectory data, and the user semantics inference techniques based on trajectory data mining from three aspects (i.e. intention, behavior and relation). The influence mechanism of user semantics on information adaptation and the interaction design approach for semantic location-aware applications will also be investigated. The final object of this project is to provide theoretical and technological basis for implementing intelligent, natural and efficient location-aware computing under pervasive computing environments and promoting its application in the domains such as electronic service, intelligent transportation, ambient intelligence and digital entertainment.