A Parallel Genetic Algorithms Framework based on

This paper describes a framework for developing parallel Ge-netic Algorithms (GAs) on the Hadoop platform, following the paradigm of MapReduce. The framework allows devel-opers to focus on the aspects of GA that are specific to the problem to be addressed. Using the framework a GA appli-cation has been devised to address the Feature Subset Selec-tion problem. A preliminary performance analysis showed promising results. Categories and Subject Descriptors C.2.4 [Computer-communication Networks]: Dis-tributed Systems—Distributed applications; D.1.