A Model for Chinese Sentence Similarity Computing
Sentence similarity computation is very important in the field of case-based machine translation. Through the in-depth analysis of sentence and the sentence similarity computing method based on the similarity computation of the word form feature, the word order feature and the semantic feature, we propose a sentence similarity computing model based on the multi-featured weight. By fusing the three features, giving different feature different weight to adapt the contribution of each feature to the sentence similarity computation, make sentence similarity computation more accurate. Experiment result shows that this approach has better accuracy in sentence similarity computation than the others.
H. Wang, B.J. Zhang, X.Z. Liu, D.Z. Luo, S.B. Zhong
X. F. Yang et al., "A Model for Chinese Sentence Similarity Computing", Advanced Materials Research, Vols. 143-144, pp. 668-671, 2011