Chinese-English cross-language inquiry antecedent expanding method based on matrix weighting association rule

Abstract

The invention discloses a Chinese-English cross-language inquiry antecedent expanding method based on matrix weighting association rules. The method includes the steps of translating Chinese inquiry into English inquiry with the aid of machine translation, retrieving an English document, conducting association judgment on primary-retrieval forefront English documents by a user to obtain a primary-retrieval association English document set, mining the matrix weighting association rules with a seccedent as the translated existing inquiry lexical item for the primary-retrieval association Englishdocument set by means of the matrix weighting association model supporting degree calculating method based on the item frequency and the weight and the matrix weighting association mode mining methodbased on the supporting degree-confidence coefficient-interestingness, and extracting an antecedent from the association rules to serve as a cross-language translated expansionw rod so that the Chinese-English cross-language inquiry translated antecedent can be expanded. It is shown through an experiment result that the problems of inquiry theme severe drifting and word mismatching existing in cross-language information retrieval for a long time can be effectively reduced, the cross-language information retrieval performance is improved, and the method has high application value and popularization prospects.

Description

technical field [0001] The invention belongs to the field of network information retrieval, in particular to a Chinese-English cross-language query antecedent extension method based on matrix weighted association rules. Background technique [0002] With the popularization of Internet technology, network information resources with multilingual characteristics have grown rapidly and become network big data with huge economic value and research value. How to retrieve information resources in other languages ​​from big data resources with familiar query language expressions to meet more information needs, and make cross-language information retrieval technology an urgently needed technology for current network users. The process of cross-language information retrieval is more complicated than that of single-language retrieval, and the problems encountered are more serious. The main manifestations are: affected by the translation quality, the query subject drifts seriously, the ...

AI Technical Summary

Problems solved by technology

[0003] The present invention proposes a Chinese-English cross-language query antecedent extension method based on matrix weighted association rules, which is applicable to the field of cross-language information retrieval, can effectively reduce problems such as query subject drift and word mismatch in cross-language information retrieval, and improve and Improve cross-language search performance

Images