面向上下文感知计算的贝叶斯网络结构自学习算法的研究

通过对上下文感知计算中上下文特点的详细分析，提出一种面向上下文感知计算的通用贝叶斯网络结构自学习方法。该方法能在足够实例数据的支撑下自动对上下文感知计算中上下文之间的关系进行学习，进而形成贝叶斯网络结构，用于从低层上下文向高层上下文的演化。通过对上下文感知计算中上下文的层次化特点的有效利用，该方法对贝叶斯网络自学习方法进行了有效优化。研究分析表明，该方法能显著降低贝叶斯网络学习过程中的时间复杂度。

Research on Bayesian network structure learning algorithm for context-aware computing

By analyzing the characteristics of context in context-aware computing, this paper presented a generic context-aware computing-oriented algorithm for Bayesian network structure learning. With sufficient data the relationships between contexts could be found by the algorithm, and a Bayesian network structure used for context inference could be formed consequently. Theoretical and experimental results all show that the algorithm can find a Bayesian network structure similar with the optimal with affordable search time.