Abstract: According to the disadvantages that traditional Web text mining technologies lack capability of ontology understanding, this paper proposes and implements a Web text mining model based on ontology. It uses a concept vector space model based on ontology instead of traditional vector space model to represent the documents. Besides, an integration information retrieval design is proposed on the foundation of text mining.

Key words: ontology, Web text mining, vector space model, information retrieval

摘要： 针对传统Web文本挖掘技术缺少语义理解能力的不足，提出并实现一种基于本体的Web文本挖掘模型，即利用基于本体概念体系的向量空间模型替代传统的向量空间模型来表示文档，在此基础上进行Web文本挖掘，并给出一种集成语义信息检索的设计。实验结果初步验证了本体模型在Web文本挖掘技术上应用的可行性。

关键词: 本体, Web文本挖掘, 向量空间模型, 信息检索

CLC Number: 

• TP393.07