Indonesian-English cross-language post-translation hybrid expansion method based on item weight sorting mining

Abstract

The invention discloses an Indonesian-English cross-language post-translation hybrid expansion method based on item weight sorting mining, which comprises the following steps of: firstly translating Indonetic language queries into English and retrieving English documents, and constructing an initial check user related feedback document set; fusing the weight and frequency of the item set with thetotal weight of the feature words and the total number of the documents of the initial detection user related feedback document set, mining the feature word frequent item set from the initial detection user related feedback document set, pruning through item weight sorting, and using the confidence-correlation coefficient evaluation framework to mining feature word weighted association rules fromfeature word frequent item sets; using an association rule antecedent item set with the posterior part being the original query lexical item and an association rule posterior part item set with the posterior part being the original query lexical item as post-translation extension words to realize Indonesian-English cross-language post-translation hybrid expansion. According to the method, the itemweight sorting pruning method is adopted, so that the mining efficiency is improved, the extension words related to the original query are mined, and the Indonesian-English cross-language text information retrieval performance is improved, and the application value and the popularization prospect are good.

Description

technical field [0001] The invention belongs to the field of information retrieval, and specifically relates to an Indonesian-English cross-language post-translation mixed expansion method based on item weight sorting and mining. Background technique [0002] Cross-language query expansion is one of the key technologies to improve and improve the performance of cross-language information retrieval. It refers to the process of cross-language information retrieval, using a certain strategy to find the expansion words related to the original query, and the combination of the expansion words and the original query is obtained. The process of re-querying and re-retrieving. According to different stages of cross-language information retrieval, cross-language query expansion can be divided into three types: query pre-translation expansion, query post-translation expansion and hybrid query expansion. The query pre-translation expansion model refers to that before the source languag...

AI Technical Summary

Problems solved by technology

[0005] The invention proposes an Indonesian-English cross-language post-translation hybrid expansion method based on item weight sorting and mining, which is applied to the field of cross-language information retrieval, applied to actual cross-language search engines and cross-language information retrieval systems, and improves cross-language retrieval performance , to solve the problem of query subject drift and word mismatch in cross-language information retrieval

Images