摘要： 提出了一种自动构造特定领域本体的方法，该方法应用术语抽取和多重聚类技术。在术语抽取阶段，通过术语在专业语料与背景语料中出现概率的对比，采用LLR公式对术语进行评分，取得了更好的抽取效果。在层级关系发现过程中，采用上下文共现信息结合HowNet中词语的语义相似度，进行术语间相似度度量，力求获得术语间最合理的相关状况。同时改进了k-medoids聚类算法，更准确地发现术语的层级关系，进而构造出特定领域的本体。

关键词: 本体, LLR, 术语抽取, 聚类, k-medoids

Abstract: This paper presents an approach to mining domain-dependent ontologies using term extraction and relationship discovery technology. There are two main innovations in the approach. One is extracting terms using log-likelihood ratio, which is based on the contrastive probability of term occurrence in domain corpus and background corpus. The other is fusing together information from multiple knowledge sources as evidences for discovering particular semantic relationships among terms. In the experiment, traditional k-mediods algorithm is improved for multi-level clustering. The approach to produce an ontology for the domain of computer science is applied and promising results are obtained.

Key words: ontology, LLR, term extraction, cluster, k-mediods

中图分类号: 

• TP391.43