An Enhanced Memetic Algorithm for Feature Selection in Big Data Analytics with MapReduce

Recently, various research fields have begun dealing with massive datasets forseveral functions. The main aim of a feature selection (FS) model is to eliminate noise, repetitive, and unnecessary featuresthat reduce the efficiency classification. In limited period, traditional FS models cannot manage filterunnecessary features. It has been discovered from state-of-the-art literature that metaheuristic algorithms perform better compared other wrapper-based techniques. Common techniques such as Genetic Algorithm (GA) andParticle Swarm Optimization (PSO) algorithm, however, suffer slow convergence local optima problems. Even new generation Firefly heuristic Fish Heuristic, these questions shown overcome. This paper introduces an improved memetic optimization (EMO) algorithm for in this perspective by using conditional criteria large datasets. proposed EMO divides entire dataset into sample blocksandconducts task learning map steps. partial result obtained combined final vector weights reductionprocess which defines appropriate collection characteristics. Finally, method grouping based on support machine (SVM) takes place. Within Spark system, applied experimental results claim it superior approaches. simulation show maximum AUC values 0.79 0.74 respectively are EMO-FS model.