机构地区: 华南师范大学计算机学院
出 处: 《计算机与现代化》 2012年第9期65-68,133,共5页
摘 要: 概念间语义相似度研究是知识表示和信息检索领域中的一个重要内容,也是自然语言处理研究的重要组成部分,是人工智能领域中一个亟待解决的问题。本文在本体的基础上,对传统的相似度计算模型进行改进,提出一种基于贝叶斯网络的概率推理方法,改进概念间语义距离的计算,从而提高了语义相似度计算模型的效果;同时采用D-分离的方法,解决了在推理过中的计算复杂性。 Concept semantic similarity is an important content in knowledge representation and information retrieval, and a critical part in natural language processing, also a problem to be solved in the field of artificial intelligence. This article is about impro- ving traditional similarity calculation model on the basis of ontology, and puts forward a kind of probability reasoning method to improve effect that is the calculation of the semantic distance between concepts, in order to perfect result about semantic similarity calculation model, meanwhile, uses D-separation to solve computational complexity in the reasoning process.
领 域: [自动化与计算机技术] [自动化与计算机技术]