Data mining techniques for proactive fault diagnostics of electronic gaming machines

Authors:
Matthew Butler;Vlado Kešelj
Affiliations:
,Faculty of Computer Science, Artificial Intelligence Group, University of York, UK;Faculty of Computer Science, Dalhousie University, Canada
Venue:
AI'10 Proceedings of the 23rd Canadian conference on Advances in Artificial Intelligence
Year:
2010

Citing 1
Cited 0

Data Mining to Predict Aircraft Component Replacement

IEEE Intelligent Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper details the preliminary research into modeling the behavior of Electronic Gaming Machines (EGM) for the task of proactive fault diagnostics The EGMs operate within a state space and therefore their behavior was modeled, using supervised learning, as the frequency at which a given machine is operating in a particular state The results indicated that EGMs did exhibit measurably different behavior when they were about to experience a fault and these relationships were modeled effectively by several algorithms.