NTT DATA at TREC-7: System Approach for Ad-Hoc and Filtering

In TREC-7, we participated in the ad-hoc task (main task) and the ltering track (sub task). In the adhoc task, we adopted a scoring method that used co-occurrence term relations in a document and speci c processing in order to determine which conceptual parts of the documents should be targeted for query expansion. In ltering, we adopted a machine-readable dictionary for detecting idioms and an inductive learning algorithm for detecting important co-occurrences of terms. In this paper, we describe the system approach and discuss the evaluation results in brief for our ad-hoc and ltering in TREC-7.