ASYMMETRIC SEMANTIC SIMILARITY FOR BIOMEDICAL INFORMATION RETRIEVAL

摘要：

　　The semantic similarity measure between concepts is an important component of text understanding, and in the last a few years, many measures have been proposed for general domain and biomedical domain. In this paper, we propose asymmetric semantic similarity measures for comparing concepts from query and document in MEDLINE document retrieval. The asymmetric measures can be built by modifying existing symmetric measures. Then the interconcept similarities are aggregated to compute the relevance score of a document. Finally, this semantic-based retrieval is combined with traditional text-based retrieval. Our approach is evaluated on OHSUMED dataset, and experimental results indicate using semantic similarity can improve MEDLINE retrieval performance, and our asymmetric semantic similarity measures are more suitable for retrieval task and achieve better performance than symmetric ones.

关键词： Biomedical Information Retrieval Asymmetric Semantic Similarity MeSH

作者: Hong Wu Kuangkai Sun

作者单位: School of Computer Science and Engineering,University of Electronic Science and Technology of China Chengdu 611731, P.R.China

会议类型: 国际会议

会议名称: The 7th Asia Pacific Association for Medical Informatics Conference(第七届亚太医药信息学大会(APAMI2012))

会议地点: 北京

会议语种:英文

页码: 1-5

在线出版日期: 2012-10-24（万方平台首次上网日期，不代表论文的发表时间）