Filters, Webs and Answers: The University of Iowa TREC-8 Results


Our existing approach to search/filtering involves a dynamic clustering technique where the threshold for formation of new clusters and the threshold for visibility of ‘sufficiently important’ clusters can be specified by the user when the topic is presented to the system. The TREC requirements for multi-query support and simulation of user judgment responses led us to modify the single set-of-clusters model, creating a two-level scheme. Note that we did not use the controlled-language field in the FT database.

Extracted Key Phrases

5 Figures and Tables

Cite this paper

@inproceedings{Eichmann1999FiltersWA, title={Filters, Webs and Answers: The University of Iowa TREC-8 Results}, author={David Eichmann and Padmini Srinivasan}, booktitle={TREC}, year={1999} }