Method and device for constructing address database

Abstract

The invention discloses a method for constructing an address database. The method comprises: an address database setting step: establishing a tree structure standard normal form address database; an address receiving step: receiving addresses to be classified; and an address classifying step: analyzing semanteme of the addresses to be classified, and classifying the addresses to be classified into the standard normal form address database according to the tree structure. The invention also discloses a semanteme analysis based address database constructing device. The invention has the advantages that the standard normal form address database is established by analyzing and classifying all received address information, and can normalize the received address information sequentially with strong controllability.

Description

【Technical field】 [0001] The invention relates to a method and device for constructing an address database, in particular to a method and device for constructing an address database based on semantic analysis. 【Background technique】 [0002] Over the past ten years, with the development of Internet technology, people have become more and more dependent on the rich, fast and timely information provided by the Internet. However, how to find the information to be searched in the vast amount of information has become an urgent problem to be solved. Correspondingly, numerous Internet search engines and corresponding websites have arisen at the historic moment, and the leader in the middle includes Baidu Search (Baidu Search ( www.baidu.com) and Google Search (www.google.cn) of Google Inc. [0003] Among the many information that needs to be searched, one important type of information is to search for address information, and this type of demand is especially paid attention to wh...

AI Technical Summary

Problems solved by technology

[0005] However, most of the existing address databases are constructed without normalization settings, and only use dictionaries and manual summarization to classify the received address data into the address database, which is manually adapted to the received address data, for example Explanation: If the received address is "No. 100 Zhongguancun Street South", it first manually adapts to the received address, and adds classification rules as road name-direction name-house number to classify the address; if the received The address is "South of No. 100, Zhongguancun Street", and the receiving address is adapted manually, and a new classification rule is added as road name-house number-location name to classify the address

However, the processing process of such a construction method is relatively complicated, and if things go on like this, continuously adapting to the receiving address manually and adding new classification rules will cause the classification rules to be disordered, scattered, and complicated. In addition, the classified address data still remains The appearance at the time of reception, the content is disordered and scattered, and it is customized to describe the phenomenon, and its controllability is poor, especially when managing the address database in the future, it will cause great trouble

Images