An Efficient Similarity Measure Algorithm of Chinese Sentence

Changying Ning,Rongbo Wang,Zhiqun Chen,Bei Lu
DOI: https://doi.org/10.1109/csae.2011.5952703
2011-01-01
Abstract:Chinese sentence similarity measures are widely used in the field of Chinese information processing. Considering practical application environment of Chat Robot and its response speed requirements, this paper presents an efficient similarity measure for Chinese sentence. In this method, a weight is assigned to the length of matched sequence strings. The longer the length of matched sequence strings, the greater the weight assigned to length, which means more similar the two compared sentences. The experimental results indicate that, compared to that of Baidu-knows, the proposed method is more reasonable and more feasible.
What problem does this paper attempt to address?