一种用于深层网接口集成的模式匹配方法

摘要

By anglicizing the limitations of existing evidence theory method for Deep Web interface integration, a Deep Web pattern matching method based on concept word and semantic heterogeneity model is proposed. The method preprocesses pattern through extracting concept word, discriminates and combines group attributes to convert m: n complex matching into 1 : 1 simple matching for improving implement efficiency. By introducing instance into semantic heterogeneity model, the problem of mining semantic heterogeneity synonymy attributes is resolved by computing, synthetic evaluating, and selecting similarity values of attribute features. Experimental results indicate that compared with evidence theory method, the efficiency and accuracy of the method is improved obviously.%针对已有证据理论(DS)方法在深层网接口集成方面的局限性,设计一种基于概念词与语义异构模型的深层网模式匹配方法.通过提取概念词对概念词模型进行预处理,识别并组合成组属性,使m∶n的复杂匹配转变为1∶1的简单匹配,提高系统执行速度.在语义异构模型中引入属性实例,将挖掘语义异构的同义属性问题,转化为对属性间各特征相似值的计算、综合评测和选取问题.实验结果表明,该方法在匹配效率和准确率上较DS方法有较大改进.