基于规则的交互上下文预测.pdfVIP

第42 瞥第7 期 Vo l.42 NO.7 天?拿大学学报 2009 年7 月 Journal Ju l. 2009 基于规则的交互上下文预测 丁春，冯志勇 (天津大学计算机科学与技术学院，天津 300072) 摘 襄:为了改进当前上下义预测中由于缺乏对交Ji.情况的考虑所导致的不究会预测问题，提出了暴于规则对交JL 上下文进行预测的方法.给出了交互上下文的定义J.l使用本体对交互上下文进行建模的方法.在基于规则的交互上下 义预测架构中，利用交互规则发掘算法和交互预测算法等推理计算得出交互规则，根据得到的交互规则对交互上下文 进行预测该方法的使用合理地改进了传统的上下义模型，从根本上系统地解决了传统的上下文预洲的局限性问题， 有效地提高了上下文预测的能力. 关键询:交豆上下文;交Ji规则: 伫下文预测 中嗣分类号: TP301 文献栋忠码:A 文章摘号: 0493-2137(2009)07嗣0628-08 Interactive-Context Prediction ßased on Rule DING Chun , FENG Zhi-yong (School of Computer Science and Technology. Tianjin University. Tianjin 300072 .China) Abstract: In ord创 to improve the inadequacy of prediction brought by the lack of considering of interaction situation in the cuηent study ofωntext prediction.an interactive-context prediction approach based on the interactive rule was proposed. The conc叩t of interactive-context and the ontology modeling method of interactive-context were introduced. In the architec- ωre of rule-based interactive帽context prediction ,the interactive rules were deduced by using interactíve rule mining algorithm and interactive prediction algorithm. The interactive-context could be predicted with the obtained interactive rules. Using this approach ,the traditional context model is improved re咽onably and the limitation of the traditional con比xt prediction is solved completely .and the ability of context prediction is improved e衍ectively. Keywords: interactive-context; interactive rule; context prediction 上下文感知川是普适计算[2) 的一个重要研究领 上述的七下文预测都是