Bulletin of Surveying and Mapping ›› 2022, Vol. 0 ›› Issue (2): 145-148.doi: 10.13474/j.cnki.11-2246.2022.0060

Previous Articles     Next Articles

Multi-strategy chinese address matching method

PENG Yulong1,2, HU Shunshi1,2, WU Tao1,2   

  1. 1. College of Geographic Sciences, Hunan Normal University, Changsha 410081, China;
    2. Key Laboratory of Geospatial Big Data Mining and Application, Hunan Province, Hunan Normal University, Changsha 410081, China
  • Received:2021-03-03 Revised:2021-06-02 Published:2022-03-11

Abstract: Address matching is a crucial link in geocoding and is one of the key technologies to realize data spatialization. Aiming at the problem that the matching rate,accuracy and time cost of the current Chinese address matching method cannot be taken into account, this study proposes a multi-strategy Chinese address matching method. The main idea is to build a lightweight dictionary for Chinese address segmentation and a multi-tree to store the address data after creating words participle. In the matching process, the fuzzy matching and hierarchical backtracking matching are combined to complete the address matching. Based on real data, this paper conducts experiments, and the results show that this method is more balanced than other matching methods in matching rate, accuracy rate and time cost.

Key words: address matching, Chinese address segmentation, multi-tree, hierarchical backtracking, cosine similarity

CLC Number: