An error correction pair mining method and system

Abstract

The embodiment of the invention discloses an error correction pair excavation method. The method comprises the steps that multiple search strings input within a preset time period are obtained; similarity calculation is conducted on each search string selected from the multiple search strings in sequence and multiple target search strings, wherein the input time of the target search strings is later than that of each selected search string; two search strings of which the similarity reaches a first preset threshold are set as a candidate error correction pair; statistical characteristics of all the candidate error correction pairs are obtained through cluster analysis, and the confidence degree of each candidate error correction pair is calculated out according to the statistical characteristics; the candidate error correction pair of which the confidence degree reaches a second preset threshold is determined as the final error correction pair. The invention further discloses an error correction pair excavation system. By means of the error correction pair excavation method and system, the technical problem that in the prior art, an input which is really needed by a user is difficult to obtain through an error correction strategy is solved, and the stability and reliability of an error correction result obtained through the error correction strategy are greatly improved.

Description

technical field [0001] The invention relates to the field of computer network communication, in particular to an error correction pair mining method and an error correction pair mining system. Background technique [0002] With the popularity of the Internet, the information resources of the Internet have grown exponentially. Search engines provide users with an important way to obtain the resources they need from the massive information resources. However, for most users, when performing an initial search, The search term (or search string) entered is relatively vague, and sometimes there are certain errors in the search term. Users cannot use more accurate search terms to search for the information they need. Only through continuous screening and refining can they obtain the final result. When the existing search engine detects the user's search term, it can provide the user with a corrected search term according to the search term entered by the user. [0003] At present...

AI Technical Summary

Problems solved by technology

[0004] However, for the error correction of the search engine, if the input A entered by the user results in zero results or very few results, the error correction string B may be directly used for retrieval. For the error correction result B given by the error correction system, whether it is the user’s It is difficult to know the real input of the current error correction strategy, which affects the stability and reliability of the error correction result; how to improve the stability and reliability of the error correction strategy for the error correction result is a hot issue that people are very concerned about

Images