基于语义相似度的领域知识推荐研究

Research on Knowledge Recommendation for Domain Ontology Based on Semantic Similarity

提出一种基于语义的领域知识推荐方法,通过判断用户输入类型,分别进行概念相似度和短句相似度的计算。其中概念相似度计算是通过计算概念的信息内容值进行的,短句相似度计算分为语义相似度和句法结构相似度。实验结果表明,该方法有效地对用户的查询请求进行概念扩充,提高了搜索的查全率与查准率。

With the explosive growth of web resource, it is difficult for keyword-based knowledge recommendation to meet the professional needs of users. In this paper, a knowledge recommandation calculation algorithm based on semantic similarity method is proposed. According to the style of user＇s input, we calculate similarity of concepts based on information content and similarity of sentences based on semantic similarity and structure similarity. Experiment results show that the user＇s inquiry request has been expanded its concept effectively, and the recall and accuracy of retrieval have been improved obviously.