提出一种新的基于本体和文档重构的语义检索方法,该方法通过构造本体知识库,依据本体知识进行文档重构,将本体的语义描述和语义关联能力应用到现有的信息检索系统。将隐形语义索引(Latent Semantic Indexing,LSI)技术应用到语义检索结果的排序过程中,并与传统的向量空间模型(Vector Space Model,VSM)方法进行对比。实验结果表明本文所提出方法更具有效性,比相对应的VSM方法性能提高约10.55%-17.63%。
To enhance the retrieval accuracy of information search engine, this paper proposes an information retrieval system based on Ontology and document refinement, which is realized by employing the semantic description and relevance of Ontology to the system. It describes the using of LSI to replace the traditional VSM in the results of sorting process. Using a comparative experiment, the authors show the new approach is more feasible and effective than VSM, which can improve the performance upto 10.55%-17.63%.
陈兵,邰晓英. 基于本体和文档重构的语义检索方法*[J]. 现代图书情报技术, 2009, 25(12): 42-46.
Chen Bing,Tai Xiaoying. Semantic Retrieval Using Ontology and Document Refinement. New Technology of Library and Information Service, 2009, 25(12): 42-46.