Feature selection architecture with parallel biogeography optimization for improved image classification in big data

Abstract

In todayand#8223;s fast-moving world, data is generated through a number newlineof sources and its fast transitioning into digital technologies is the main reason newlinefor the development of big data. Another open-source software framework newlinewhich allows huge datasets is Apache Hadoop which can be processed with newlinethe help of commodity hardware. It is designed in such a way that it can run newlineon top of large clusters of nodes which are connected with one another in newlineforming a huge distributed system. MapReduce is the paradigm implemented newlineby Hadoop which gains its inspiration from a design developed by Google in newlineimplementing its search technology. The Hadoop Distributed File System newline(HDFS) indicates the component of the Hadoop and its storage that has been newlinedesigned to store large datasets on clusters in order to stream them to the other newlineuser applications with higher throughput newline