Evaluating Resources for Query Translation in Cross-Language Information Retrieval

Our goal is to evaluate the utility of a lexical resource containing Lexical Conceptual Structures (LCS) for use in cross-language information retrieval. Our evaluation makes use of a combination of techniques from interlingual machine translation with conventional information retrieval techniques. Given a query in one language, we transform the query into the corresponding terms in a second language. Our focus is on the construction of disambiguated targetlanguage queries by using verb-based entries in our lexicon to construct Lexical Conceptual Structures. The main innovation of LCS-based query translation is that it provides a principled framework for controlling translation ambiguity in cross-language information retrieval applications. We evaluate this approach by comparing the resulting retrieval effectiveness with alternative techniques based that use offthe -shelf machine translation or translation knowledge extracted from machine readable dictionaries. We view this work as an important demonstration of techniques for evaluating the utility of alternative structures for multilingual lexicons in cross-language IR applications. In addition, this work provides a basis for measuring the extent to which disambiguation can enhance cross-language information retrieval effectiveness.