基于Web本体挖掘的语义目录研究

Semantic-catalogue research based on web ontology-mining

针对现有Web资源访问模式缺乏针对性、信息冗余、缺乏语义等缺点,提出一种区别于传统Web结构的新的目录概念——语义目录,对目录的生成方法提出了解决方案。利用类Apriori算法对用户日志进行挖掘得到频繁页面规则集。本体Agent对规则集进行提取,得到的本体元和用户模式分别存储于本体知识库和频繁路径序列模式树(FRSP-tree)中,并且在FRSP-tree树结点中加入指向本体元的指针,使遍历FRSP-tree树生存的目录具有语义性和针对性。

Now the access model of the web resource is unsematic and has no pertinence, information redundancy. In order to change these disadvantages, a new catalog notion-semantic catalog which is different from the traditional web structure is put forward. A similar Apriori algorithm is proposed for getting the frequence-sets. Then pu.t the result into the FRSP-tree which includes the ontology. Search in the FRSP-tree and match the pattern of usage to form the semantic-catalogue.