基于本体的文档信息检索模型优化

For the traditional document retrieval model based on keyword matching, there is a problem that the query keyword must exactly match the document index keyword, and the problem that the document containing the synonym relationship and the inclusion relationship with the search keyword cannot be retrieved is raised. The document similarity matching phase introduces an improved document information retrieval model that contains ontology in the same domain as the query input sentence. In order to solve the problem that the query keyword must exactly match the document index keyword, it is proposed that the query includes the ontologies in the same domain as the query input sentence to query and expand the query input keyword. In order to solve the problem that it is impossible to retrieve the existence of a synonymous relationship and a relational document containing search keywords, an improved vector space model based on domain-specific ontology is proposed for document similarity matching. Experiments show that by using an improved ontology-based document information retrieval model, the recall rate and accuracy of the document can be significantly improved. In particular, for the polysemy keywords, the recall rate is particularly noticeable.