Semi-automatic construction of the Chinese-English MeSH using Web-based term translation method

Area of Science:

• Medical Informatics
• Natural Language Processing
• Information Retrieval

Background:

• Language barriers limit access to U.S. medical information for non-English users.
• Existing cross-language medical information retrieval (CLMIR) systems often rely on multilingual thesauri like MeSH.
• MeSH lacks a current traditional Chinese translation, hindering CLMIR for Chinese speakers.

Purpose of the Study:

• To propose a semi-automatic approach for constructing a Chinese-English MeSH thesaurus.
• To address the absence of a traditional Chinese MeSH translation.
• To facilitate improved CLMIR for Chinese-speaking users.

Main Methods:

• A semi-automatic approach utilizing web-based term translation was employed.
• Candidate terms were mined from anchor texts and search results.
• Knowledge engineers utilized these mined terms for thesaurus construction.

Main Results:

• Over 19,000 Chinese-English MeSH entries have been compiled.
• The developed thesaurus is ready for use in future CLMIR applications.
• The semi-automatic method proved effective in generating a substantial bilingual thesaurus.

Conclusions:

• The constructed Chinese-English MeSH thesaurus is a valuable resource.
• This work significantly enhances the potential for Chinese-English CLMIR.
• Future applications will leverage this thesaurus to bridge language gaps in medical information access.