计算机技术与发展Issue(12)：37-40,4.DOI:10.3969/j.issn.1673-629X.2014.12.009
多特征结合的词语相似度计算模型
Word Similarity Computation Model of Multi-features Combination
摘要
Abstract
Semantic similarity computing has been widely used in machine translation based on example,information retrieval and auto-matic question answering systems. Word similarity computation is generally based on the original in "HowNet",through calculating the degree of similarity between concepts to obtain. In this paper,in consideration of the original distance,depth,width,density and contact ratio,use the method with multi-features to compute word similarity. In order to verify the rationality of the algorithm,using the bench-mark of words given by Miller and Charles literature as a test set,make a comparison between the word similarity computation values and expert value,calculating the Pearson correlation coefficient,the calculation results is 0. 852. Experimental result show that the word simi-larity computation of multi-features combination is identical with expert estimation.关键词
词语相似度/知网/同义词词林/语义距离Key words
word similarity/HowNet/Tongyici Cilin/semantic distance分类
信息技术与安全科学引用本文复制引用
张培颖,房龙云..多特征结合的词语相似度计算模型[J].计算机技术与发展,2014,(12):37-40,4.基金项目
中央高校基本科研业务费专项资金(13CX02031A) （13CX02031A）
