Method and system for retrieving organization names

Abstract

The invention relates to a method and a system for retrieving organization names. The method comprises the following steps of: automatically segmenting all organization names into place names R, identifications I, types T and suffixes S; converting all the organization names into abbreviated letters; establishing a full-text index and language model corresponding to letter abbreviations of all the identifications I; establishing a place name dictionary, a type synonym dictionary and a suffix synonym dictionary; receiving an input letter string and segmenting into identifications I for full-text retrieval to retrieve the place names R, the types T and the suffixes S so as to obtain candidate retrieval results; and sequencing and outputting. The invention can be used for accurately and quickly retrieving Chinese and English organization names by using letters to realize the furthest fault tolerance of user inputs, shortens the information search time on the premise of no accuracy loss and also has the automatic learning function.

Description

【Technical field】 [0001] The invention relates to a search method, in particular to a search method and system for an organization name. 【Background technique】 [0002] The rapid query technology of massive organization names is an important tool for operators to improve work quality and work efficiency in voice services. At present, such tools are widely used in Telecom 114 directory inquiry platform and China Mobile 12580 information service platform. On the basis of the huge business data in the background of Number Bestone, [0003] Operators must provide users with the most accurate information in the shortest possible time, but no operator can grasp the vast and boundless business data. Therefore, the institution name quick query system is an indispensable tool for operators to provide services to users. Most of the current query systems are expanded based on the Pinyin coding specification. The operator inputs the Pinyin letters according to the coding specificatio...

AI Technical Summary

Problems solved by technology

[0004] (1) The coding is rigid and the fault tolerance is poor

Because the operator does not have all the information, they cannot tell whether the name of the institution stated by the querying customer is incorrect or incomplete

At present, the system cannot provide fuzzy processing for the "wrong" query input by the operator, which eventually results in the service result of "no such person found".

In fact, it is very easy to make small mistakes in the correc

Images