Feature Selection in MicroArray Datasets using Manta Ray Foraging Optimization Technique

Abstract

In this document, we apply the Manta Ray Foraging Optimization technique for Feature Selection in the genetic dataset to identify the relevant features from a high dimensional feature space. MRFO is one of the recent meta-heuristic optimization techniques which has been used for Feature Selection. presence of cancerous tumors in the patient. [10]. This document analyzes the pre existing algorithms developed in the space of Feature Selection using the MRFO algorithm in the GEM dataset. The MRFO method uses three unique search techniques in three different ways. The local search ability of the algorithm is highly aided by the chain foraging behavior; the global search ability is determined by the manta ray cyclone foraging behavior; and the local search ability and convergence rate are greatly improved by the somersault foraging behavior. This document, two separate learning classifiers were used: K-NearestNeighbors (KNN) classifier and the Random Forest Classifier were used for calculating the fitness