中文摘要：

上下文信息的缺失是上下文信息处理中不可避免的问题,缺失数据插补方法也是数据挖掘中的研究热点。但是,现有的缺失数据的插补方法不太适合上下文信息这一流数据形式,没有充分利用各传感器采集数据之间的关联性,而且在插补的过程中没有考虑传感器数据的时空关系。为了解决现存的缺失数据插补方法的缺陷和不足,该文提出了基于时空关系和关联规则挖掘的上下文信息缺失插补方法（STARM）,对传感数据进行空间化和时间序列化,并生成强关联规则对缺失数据进行插补。最后,通过温度传感器采集数据验证了这一算法合理性和高效性。实验证明,该算法在上下文信息缺失估计的准确性要高于简单线性回归算法（SLR）和EM算法等,而且具有较小的时空开销,能够保证实时应用的服务质量（QoS）。

英文摘要：

The context data missing is an inevitable problem of context information processing mechanism, the imputation technique of missing data also is a research hotspot in data mining. However, the existing imputation technique of missing data is not suitable for the flow data form of context information that does not make full use of data relevance between every collecting sensor. Moreover, that does not take spatial-temporal relationship into account. In order to conquer the shortcomings and deficiencies of the existing imputation technique of missing data, this paper proposes an imputation technique for context data missing based on Spatial-Temporal and Association Rule Mining （STARM） to perform spatiality and time series analysis on sensor data, and generate strong association rules to interpolate missing data. Finally, the simulation experiment verifies the rationality and efficiency of STARM through temperature sensor data acquisition. Experiments show that the algorithm is of high accuracy for the imputation of context data missing, such as Simple Linear Regression （SLR） algorithm and the EM algorithm. In addition, which is smaller time and space overhead and can guarantee Quality of Service （QoS） of real-time applications.