Fast Evaluation Techniques for Complex Similarity Queries

Title Fast Evaluation Techniques for Complex Similarity Queries
Author(s) K. Böhm, M. Mlivoncic, H.-J. Schek, R. Weber
Type Article
Booktitle 27th Int. Conf. on Very Large Databases (VLDB)
Roma, Italy
Organization Institute of Information Systems, ETH Zurich
Month September
Year 2001


Complex similarity queries, i.e., multi-feature multi-object queries, are needed to express the in-formation need of a user against a large multi-media repository. Even if a user initially issues a single-object query over one feature, a system with relevance feedback will automatically gener-ate a complex similarity query. Relevance feed-back is only useful if response times are inter-active. Therefore, this article contributes to the important problem how to evaluate such complex queries efficiently. We describe a new evalua-tion technique called Generalized VA-File-based Search (GeVAS). It builds on the VA-File, supports queries over several feature types, and borrows the idea to search an index structure with several query objects in parallel from Ciaccia et al. Our main contributions are twofold: 1) we show that GeVAS does not degenerate for queries with many objects or many feature types. 2) We develop a number of variants of GeVAS, tailored to the different distance measures and distance-combining functions, and we show that they yield a significant performance improvement.

You can directly download a PDF (144 KB) version of this paper.
!!! Dieses Dokument stammt aus dem ETH Web-Archiv und wird nicht mehr gepflegt !!!
!!! This document is stored in the ETH Web archive and is no longer maintained !!!