基于概念模型的个性化搜索引擎研究

在对用户兴趣模型探讨的基础上,提出了一种基于概念的用户兴趣模型,用于区别用户兴趣的大小.讨论了基于链接的查询聚类算法,并针对该算法的不足提出了一种基于概念的聚类算法,该算法根据用户兴趣模型建立查询-概念二分图,然后计算图中查询顶点间的概念相似度,并将概念相似度最高的查询顶点进行合并以实现聚类.设计实现了一个基于Web数据挖掘的个性化搜索引擎系统,对系统的个性化查询进行了测试,并对比分析了链接聚类和概念聚类的实验结果.

Personalized Search Engine Based on Conceptual Model

Based on discussion of user interest model, a new user model based on concept is proposed to distinguish the user＇s interest. Then a query clustering algorithm based on hyperlink is discussed in detail. For the problem of hyperlink query clustering algorithm, a new query clustering algorithm based on concept is proposed. In this clustering algorithm, query-concept bipartite graph are created through user model, and then the similarity between query vertex and concept vertex is calculated, the algorithm is itera- tive until the max similarity meets certain condition. A personalized search engine based on Web data mining is designed and imple- mented to evaluate the effectiveness of concept query clustering algorithm, and an experiment is conducted to compare with link query clustering algorithw~ Experiment shows the precision and recall of personalized search.