Title: An Alternative Implementation of the Apriori Algorithm

When: Monday, September 20, 2004 12:00 noon

Where: Room 501, IS Building

Who: Denis L. Nkweteyim, PhD Candidate, Information Science Program

Abstract: The apriori algorithm has been instrumental in the mining of association rules from a transactions database. The classical implementation of the algorithm has 2 main drawbacks: (1) the number of candidate itemsets generated by the join step may be too large to fit into main memory, and (2) high latency resulting from the need to scan the database during each iteration.

In this talk, I will present an alternative implementation of the apriori algorithm that applies the apriori property to each record in the transactions database, thus eliminating the join step. I will also present results of a simulation study that compares the performance of the new algorithm to the classical one.

Speaker Bio
*************
Denis Nkweteyim is a Ph.D. candidate in the Information Science program of the School of Information Sciences, University of Pittsburgh. He holds a BS degree in Electrical & Electronic Engineering from the University of Wales, Cardiff, and an MS degree in Digital Communication Systems from Loughborough University. His current research interests are in the area of personalization systems.