Thomson Legal and Regulatory at NTCIR-5: Japanese and Korean Experiments

Thomson Legal and Regulatory participated in the CLIR task of the NTCIR-5 workshop. We submitted formal runs for monolingual retrieval in Japanese and Korean, as well as for bilingual English-to-Japanese retrieval. We employed enhanced tokenization for our Japanese and Korean runs and applied a novel selec-tive pseudo-relevance feedback scheme for Japanese. Our bilingual search participation was a straightfor-ward application of an off-the-shelf Machine Trans-lation system to transform an English query into a Japanese query. Unfortunately we cannot draw many conclusions from our participation, as our experiments were ham-pered by technical difficulties, particularly with our to-kenization and stemming components