English-Japanese Cross-lingual Query Expansion Using Random Indexing of Aligned Bilingual Text Data

Vector space models can be used for extracting semantically similar words from the co-occurrence statistics of words in large text data. In this paper, we report on our NTCIR 2002 experiments using the Random Indexing vector space method for extract-
ing an English-Japanese cross-lingual thesaurus from aligned English-Japanese bilingual data. The cross-lingual thesaurus has been used for automatic cross-lingual query expansion in the NTCIR patent retrieval task.