Efficient profile matching for large scale Webcasting

Abstract

This paper presents an efficient method of matching diverse data against user profiles in large-scale Webcasting systems. Its design and implementation are described in the context of the Grand Central Station (GCS) project at IBM Almaden Research Center. Initial performance evaluation indicates the ability of GCS profile matching to scale up and achieve Strong performance via dynamic adaptation. © 1998 Published by Elsevier Science B.V. All rights reserved.