GEODESIC DISTANCE BASED APPROACH FOR SENTENCE SIMILARITY COMPUTATION

摘要：

This paper presents a novel approach based on Geodesic distance for sentence similarity computation, which can be used in a query-based information retrieval system. Unlike the traditional distance methods, Geodesic distance takes into account the spatial relationships of sentences, which better reflects the intrinsic geometric structure of sentence manifold. Experiments demonstrate that the proposed method shows a better correlation to human intuition compared with traditional Euclidean method.

关键词： Geodesic distance Euclidean distance Cosine angle distance Sentence similarity Manifold.

作者: HUI-FANG MA QING HE ZHONG-ZHI SHI

作者单位: The Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese The Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese

会议类型: 国际会议

会议名称: 2008 International Conference on Machine Learning and Cybernetics(2008机器学习与控制论国际会议)

会议地点: 昆明

会议语种:英文

页码: 2551-2557

在线出版日期: 2008-07-12（万方平台首次上网日期，不代表论文的发表时间）