Research Of Sentence Similarity Computation Based On Semantic Analysis

Similarity computation in such aspect as Chinese word and sentence is an essential task and widely used in the Chinese information processing.It can decide the development of related research directions somewhat.For example,in the area of EMBT,information retrieval,automatic question-answering,spelling and data cleansing,how to compute the similarity is one of the most important problem which is also a hotspot and very difficulty that people study for a long time.in this paper,we study the similarity computing of Chinese words and sentences separately.The current popular words similarity computing based on Hownet is improved by study on current words similarity computing algorithm.By studied on the algorithm of sentences similarity computing and their defects,the sentences similarity computing based on semantic analysis is presented.A simplified words dictionary for facilitating the semantic analysis are set up in experiment and this method is proved to be effective by comparison in experiments.