Abstract: An ontology based information filtering method is proposed in this paper. This approach describes formal semantic through ontology so that user’s original input query terms can be converted into ontology semantic extension formats with constraint rules. Furthermore, semantic vector definition of information and its right computation method are presented in order to achieve the semantic matching between user’s requirements and information. Finally, semantic similarity computation based information filtering is addressed. Experiments show that our information filtering algorithm is effective.

Key words: ontology, semantic extension, constraint, information filtering, similarity computation

摘要： 提出一种基于本体的信息过滤方法。该方法通过本体实现形式化语义描述，并对原始输入条件进行带约束规则的本体语义扩展。进而为了实现语义匹配，给出了信息向量语义描述及权重计算方法。最终，实现基于语义相似度计算的信息过滤。实验证明，该方法是有效的。

关键词: 本体, 语义扩展, 约束, 信息过滤, 相似度计算