Exploiting Parallel Processing Power of GPU for High Speed Frequent Pattern Mining
Department of Computer Science & IT, Mahatma Gandhi University, Meghalaya, India
International Journal of Computer Engineering and Applications, Volume VII, Issue II, 2014
@article{albert2014exploiting,
title={Exploiting Parallel Processing Power of GPU for High Speed Frequent Pattern Mining},
author={Albert, D William and Fayaz, K and Babu, D Veerabhadra},
year={2014}
}
Frequent pattern mining is one of the widely used data mining techniques for discovering trends or patterns from databases. As data is growing in exponential pace, data mining activities need more powerful computing. Fortunately modern GPUs (Graphics Processing Units) have specialized electronic circuits and support parallel processing. GPUs are capable of processing huge amount of data in short span of time. Data mining algorithms cannot be used directly with GPU. In this paper we proposed a new algorithm known as GPU Based Frequent Pattern Mining (GFPM) for high speed frequent pattern mining. The algorithm has additional mechanisms to handle memory and data transfer between traditional CPU and GPU. General Purpose Graphics Processing Unit (GPGPU) platform OpenCL with Java extensions is used to implement the algorithm. We built a prototype application to demonstrate the proof of concept. Datasets are obtained from UCI machine learning repository. The empirical results reveal that the GPU based FPM algorithm is able to leverage parallel processing power of GPU.
August 23, 2014 by hgpu
