A method for matching Chinese place-name data
Conversion and sharing of spatial data from different departments is an essential part of information construction in China.The first step of the solution is to match place-name data.However,there are administrative changes in some places with the development of urbanization process.It undoubtedly increases the difficulty to match place-name data.In the daily work,the data are artificially matched with available place-name database and materials such as graphs and record cards.Although it is easy to put in practice,this method may cost a lot of time and labor to keep the accuracy.The algorithms for matching strings can be used to solve the problem.But most of them focus on solving the English strings match problems and less refer to Chinese.In the paper,BPM-BM (Bit-Parallel Matrix -Boyer Moore) algorithm,the most efficient filter method for approximate string matching of Chinese text,is proposed to match place-names between the national surveillance sites of infectious diseases and the 1:1,000,000 scale township map of China in 2000.The study indicated that the proposed method decreased artificial process greatly and the accuracy which achieved 94.2％ was higher than the SQL commands method.
Place-name cartology edit distance Chinese approximate string matching bit-parallelism filtering BPM BPM-BM SQL commands
L.Yilan W.Jinfeng
Institute of Geographical Sciences and Natural Resources Research,Chinese Academy of Sciences,11 Dat Institute of Geographical Sciences and Natural Resources Research,Chinese Academy of Sciences,11 Dat
国际会议
第16届国际地理信息科学与技术大会(16th International Conference on GeoInformatics and the Joint Conference)
广州
英文
2008-06-28（万方平台首次上网日期，不代表论文的发表时间）