Query Translation by Text Categorization

Query Translation by Text Categorization Patrick Ruch SIM, University Hospital of Geneva 24 Micheli du Crest 1201 Geneva, Switzerland and LITH, Swiss Federal Institute of Technology 1015 Lausanne, Switzerland patrick.ruch@sim.hcuge.ch Abstract We report on the development of a cross lan- guage information retrieval system, which translates user queries by categorizing these queries into terms listed in a controlled vo- cabulary. Unlike usual automatic text cat- egorization systems, which rely on data- intensive models induced from large train- ing data, our automatic text categorization tool applies data-independent classifiers: a vector-space engine and a pattern matcher are combined to improve ranking of Med- ical Subject Headings (MeSH). The cate- gorizer also benefits from the availability of large thesauri, where variants of MeSH terms can be found. For evaluation, we use an English collection of MedLine records: OHSUMED. French OHSUMED queries - translated from the original English queries by domain experts- are mapped into French MeSH terms; then we use the MeSH con- trolled vocabulary as interlingua to trans- late French MeSH terms into English MeSH terms, which are finally used to query the OHSUMED document collection. The first part of the study focuses on the text to MeSH categorization task. We use a set of MedLine abstracts as input documents in order to tune the categorization system. The second part compares the performance of a machine translation-based cross lan- guage information retrieval (CLIR) system with the categorization-based system: the former results in a CLIR ratio close to 60%, while the latter achieves a ratio above 80%. A final experiment, which combines both approaches, achieves a result above 90%. 1 Introduction Cross Language Information Retrieval (CLIR) is increasingly relevant as network-based re- sources become commonplace. In the med- ical domain, it is of strategic importance in order to fill the gap between clinical records, written in national languages and research re- ports, massively written in English. There are several ways for handling CLIR. Historically, the most traditional approach to IR in gen- eral and to multilingual retrieval in particular, uses a controlled vocabulary for indexing and retrieval. In this approach, a librarian selects for each document a few descriptors taken from a closed list of authorized terms. A good ex- ample of such a human indexing is found in the MedLine database, whose records are man- ually annotated with Medical Subject Headings (MeSH). Ontological relations (synonyms, re- lated terms, narrower terms, broader terms) can be used to help choose the right descriptors, and solve the sense problems of synonyms and homographs. The list of authorized terms and semantic relations between them are contained in a thesaurus. A problem remains, however, since concepts expressed by one single term in one language sometime are expressed by dis- tinct terms in another. We can observe that terminology-based CLIR is a common approach in well-delimited fields for which multilingual thesauri already exist (not only in medicine, but also in the legal domain, energy, etc.) as well as in multinational organizations or coun- tries with several official languages. …

Content truncated to 500 words