Uniform Indexing and Retrieval Scheme for Chinese, Japanese, and Korean

This paper reports on our work at the third NTCIR workshop on the subtasks of Chinese, Japanese, and Korean monolingual information retrieval (IR). A Chinese IR system is applied to all document sets in these three languages. Based on the n-gram indexing model and a phrase formulation method to extract longer key terms for indexing, no language-dependent modifications were made to apply the system to Japanese and Korean IR. Our attempt is to see whether such a system originally designed for Chinese IR can still work for Japanese or Korean documents. The results turn out that it performs similarly among the document sets in these three different languages.