Cross language information retrieval : a research roadmap

Cross-Language Information Retrieval (CLIR) has been a research sub-field for more than a decade now. The field has sparked three major evaluation efforts: the TREC Cross Language Track which currently focuses on the Arabic language, the Cross-Language Evaluation Forum (CLEF) - a spinoff from TREC - covering many European languages, and the NTCIR Asian Language Evaluation (covering Chinese, Japanese and Korean). During this one-day workshop we reviewed and assessed the progress that has been made so far and discussed what research and development remains to be done to make CLIR a practical enterprise. Presentations focused on the major techniques and accomplishments of the field (e.g. utilization of corpus, dictionary, and machine translation techniques for crossing language barriers, strategies for sense disambiguation and query expansion) and position papers suggested the directions that research should take in the next half decade. The goal of our workshop was to develop a step-by-step, year-by-year roadmap of research to be undertaken, with each year addressing progressively more difficult goals and expected accomplishments.