CONCEPTUAL RETRIEVAL BASED ON FEATURE CLUSTERING OF DOCUMENTS Youjin Chang + , Ikkyu Choi + , Jongpill Choi + , Minkoo Kim + and Vijay V. Raghavan # +(xaritas, ikchoi, cjp, minkoo)@ ajou.ac.kr Dept. of Information & Computer Engineering, Ajou University San5, Wonchun-dong, Paldal-gu, Suwon, 442-749, Korea #raghavan@cacs.louisiana.edu The Center for Advanced Computer Studies University of Louisiana at Lafayette, Lafayette, LA 70504, USA Abstract In the Web search, since users' queries usually consist of only a few words, it is hard to identify their information needs. To solve this problem, many approaches have been tried to expand initial queries and to reweight the terms in the expanded queries using users' relevance judgments. Although relevance feedback is most effective when relevance information about retrieved documents is provided by users, it is not a fully automatic method. Another solution is to use correlated terms for query expansion. The main problem with this approach is how to construct the term-term correlations that can be used effectively to improve retrieval performance. In this study, we try to construct query concepts that denote users' information needs from a document space, rather than to reformulate initial queries using the term correlations and/or users' relevance feedback. To make query concepts, we extract features from each document, and cluster the features into primitive concepts that are used to form query concepts. Experiments are performed on a TREC collection. Key words Concept-based retrieval, query concepts, document features, feature clustering