Applying a Hybrid Query Translation Method to Japanese/English Cross-Language Patent Retrieval

TL;DR

This paper presents a hybrid query translation approach for Japanese/English cross-language patent retrieval, utilizing multiple dictionaries and collocational statistics to improve translation accuracy and retrieval performance.

Contribution

It introduces a novel hybrid translation method combining dictionary and statistical data, demonstrating improved retrieval effectiveness over simple dictionary-based methods.

Findings

Achieved 76% of monolingual retrieval precision

Outperformed simple dictionary-based translation methods

Validated effectiveness on Japanese/English patent abstracts

Abstract

This paper applies an existing query translation method to cross-language patent retrieval. In our method, multiple dictionaries are used to derive all possible translations for an input query, and collocational statistics are used to resolve translation ambiguity. We used Japanese/English parallel patent abstracts to perform comparative experiments, where our method outperformed a simple dictionary-based query translation method, and achieved 76% of monolingual retrieval in terms of average precision.