Analysis of the Usage of Japanese Segmented Texts in NTCIR Workshop 2.
Masaharu Yoshioka, Kazuko Kuriyama, Noriko Kando · 2001
In this paper, we report on the usage of Japanese segmented texts and analyze the submitted search results to NTCIR Workshop 2, which used these texts. In these texts, each sentence is segmented into terms and term components (similar to phrases and words). However, the sizes of terms are inconsistent in the texts; e.g., some terms that should be decomposed into term components remain as terms. We analyze the effect of this inconsistency from the viewpoint of comparison between word-based indexing and phrasal indexing. Based on this analysis, we propose the desired specification of a morphological analyzer for Information Retrieval.