摘要
为了能更为准确地计算两个语句之间的语义相似度;提出了一种基于分段语义比较的语句相似度计算方法。将语句分成主干部分和修饰部分,并根据设定的语法规则,对修饰部分进行更为细致的划分;计算语义相似度时,将计算过程分为两部分:主干部分计算、修饰部分计算,并赋以不同权值;能够按照语法和语义结构,更为合理、细致地计算出两个语句之间的语义相似程度。实验结果表明了该方法的能够取得较好的计算结果。
In order to compute sentences similarity more accurately, a sentence similarity computing method based on segmented semantic comparison is proposed. A sentence is divided into segments by general grammar rules. The trunk segment and the other segments are set different weights in sentence similarity computing, the more accurate and reasonable similarity between two sentences is calculated according to the syntactic and semantic structure of the sentences. Finally an experiment is provided to verify the effectiveness of the method.
出处
《计算机工程与设计》
CSCD
北大核心
2013年第7期2637-2640,F0003,共5页
Computer Engineering and Design
基金
国家自然科学基金项目(U1204402)
关键词
语义比较
语句分段
相似度计算
主干相似度
语句相似度
semantic comparison
sentence segmentation
similarity computing
trunk similarity
sentence similarity
