Cross-lingual post-translation hybrid extension method based on feature word weighted association pattern mining

Abstract

The invention discloses a cross-language post-translation mixed extension method based on feature word weighted association mode mining. First, the source language query retrieves the target language document for the first time, constructs and preprocesses the initial inspection related feedback document set, and compares the document by item set weight Mining the frequent item sets containing the original query items, pruning the candidate item sets with the item set correlation degree and the item weight of the item set or the largest item weight value, and using the chi-square analysis-confidence evaluation framework to analyze the frequent item sets Mining the association rules of text feature words containing the original query term, extracting the subsequent part is the association rule antecedent of the original query term and the antecedent is the association rule of the original query term, and the latter part is used as the post-translation expansion word to realize cross-language post-translation Hybrid extensions. The present invention overcomes the defects of the existing weighted association rule mining technology, improves the mining efficiency, digs out the expansion words related to the original query, improves and improves the cross-language retrieval performance, and has high application value and Promote prospects.

Description

technical field [0001] The invention belongs to the field of information retrieval, in particular to a cross-language post-translation mixed extension method based on feature word weighted association mode mining. Background technique [0002] Cross-language query expansion is one of the core technologies to improve and improve the performance of cross-language information retrieval. It can solve the long-term problems in cross-language information retrieval, such as serious drift of query topics and word mismatches. , using a certain strategy to find the expansion words related to the original query, the expansion words and the original query combination to get a new query and re-retrieval process. [0003] At present, the rapid growth of network information resources has become network big data with huge economic value and research value. Faced with network information resources with multilingual characteristics, when network users use query expressions in languages ​​the...

AI Technical Summary

Problems solved by technology

[0004] The invention proposes a cross-language post-translation mixed extension method based on feature word weighted association mode mining, which is applied to the field of cross-language information retrieval, solves the problems of query topic drift and word mismatch in cross-language information retrieval, and is suitable for actual cross-language search Engine and web cross-language information retrieval system to improve information retrieval performance

Images