Administrative Divisions of Addresses Matching Algorithm Based on Moving Window Algorithm for Maximal Matching

摘要

Maximum matching algorithm which commonly used in Chinese word segmentation algorithm is limited by word length settings, it results in the decrease of recognition efficiency. In order to improve the matching efficiency of Chinese word segmentation, in this paper, administrative divisions of address matching algorithm based on moving window algorithm for maximal matching is proposed on the basis of Chinese address word segmentation. According to semantic features of Chinese address data expression, the algorithm establishes address matching rules, finds out all administrative divisions in the address data, and achieves Chinese address segmentation database match at the same time. For 380,000 address data, maximum matching method and administrative divisions of address matching algorithm based on moving window algorithm for maximal matching match the standard database and segment Chinese address respectively in the paper. Experiments show that the algorithms proposed in this paper have shown improvement comparing with the current maximum matching method in the segmentation accuracy of administrative divisions. The algorithm realized the function that two steps of the address word segmentation and database match working at the same time to verify the validity of the algorithm.