EuroMUG '03 -- 9 - 10 Oct, 2003

Data Fusion of Similarity Coefficients for Improved Database Searching and Diversity

Jenny Chen
Daylight Research Student
Department of Information Studies, University of Sheffield
Regent Court, 211 Portobello Street, Sheffield, S1 4DP, UK


Clustering techniques have been applied to the results of similarity searches, using selection of similarity coefficients, and a subset of complementary coefficients has been identified. These coefficients have then been systematically combined using data fusion techniques in order to improve searching performance. It was apparent from the results that the choice of best performing combination is dependent on the size of the query and the size of the compounds within the database. We have carried out further evaluations and mathematical modeling in order to examine this relationship further.

Presentation slides:

Daylight Chemical Information Systems, Inc.