DocumentCode
2243094
Title
Study on the standardized method of Chinese addresses based on expert system
Author
Bin Gu ; Yanfeng Jin ; Chang Zhang
Author_Institution
Shijiazhuang Posts & Telecommun. Tech. Coll., Shijiazhuang, China
fYear
2012
fDate
Oct. 30 2012-Nov. 1 2012
Firstpage
1254
Lastpage
1258
Abstract
Structured text is the important substance for Chinese processing, including words dividing, drawing of the key words etc. which differs greatly from English. Among industry applications, Chinese phrases, especially the issues of address processing and standardization, need large number of experts´ experience, and the efficiency is very low. At the same time, this problem is the core problem based on address integration of customers´ data. Being the most important application field, Expert system contains a huge supply of knowledge in the field of expertise. Based on the analysis of structured address data method, a standardized method of addresses based on expert system was proposed in this paper. The method can process address error fuzzily and process the confused Chinese addresses into complete standardized addresses that are suitable for exact match, it also can achieve self-learning ability by updating knowledge base.
Keywords
expert systems; natural language processing; text analysis; unsupervised learning; Chinese address standardized method; Chinese phrases; Chinese processing; English language; customer data integration; expert system; industry applications; key word drawing; knowledge base update; self-learning ability; structured address data method analysis; structured text; word division; Cities and towns; Expert systems; Floors; Standards; Writing; chinese words dividing; edit distance; expert system; keyword drawing; text mining;
fLanguage
English
Publisher
ieee
Conference_Titel
Cloud Computing and Intelligent Systems (CCIS), 2012 IEEE 2nd International Conference on
Conference_Location
Hangzhou
Print_ISBN
978-1-4673-1855-6
Type
conf
DOI
10.1109/CCIS.2012.6664585
Filename
6664585
Link To Document