Abstract

FULGUEIRA CAMILO, Marlis  and  REYES BURUNATE, William. Fuzzy Miner algorithm parallelization with OpenCL. Rev cuba cienc informat [online]. 2016, vol.10, n.4, pp.1-14. ISSN 2227-1899.

ABSTRACT Process mining is a discipline that provides a set of algorithms that facilitate obtaining information from event logs. However, the processing of large amounts of data sometimes costly, due to the execution times dedicated to this task. This article reflects an implementation of the algorithm Fuzzy Miner mining process using the C ++ language. The implemented algorithm is based on a proposal made in the second article, in which parallel programming techniques are used, specifically, OpenCL. The objective pursued with this study is to reduce the processing time of the algorithm implemented Fuzzy Miner. The tests carried out using different data and different hardware platforms, thus significantly reducing runtime while processing the data in parallel. The most relevant results are obtained with the records of major events and the best architectures.

Keywords : Miner Fuzzy; OpenCL; Parallel Computing.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )