Scalable Feature Selection in High-Dimensional Data Based on GRASP

Abstract

Feature selection in high-dimensional data is one of the active areas of research in pattern recognition. Most of the algorithms in this area try to select a subset of features in a way to maximize the accuracy of classification regardless of the number of selected features that affect classification time. In this article, a new method for feature selection algorithm in high-dimensional data is proposed that can control the trade-off between accuracy and classification time. This method is based on a greedy metaheuristic algorithm called greedy randomized adaptive search procedure (GRASP). It uses an extended version of a simulated annealing (SA) algorithm for local search.In this version of SA, new parameters are embedded that allow the algorithm to control the trade-off between accuracy and classification time. Experimental results show supremacy of the proposed method over previous versions of GRASP for feature selection. Also, they show how the trade-off between accuracy and classification time is controllable by the parameters introduced in the proposed method.

Ensembles for feature selection: A review and future trends

“…• Scalability measures. As said above, data is becoming larger increasingly, in both samples and feature dimensions, a fact that at the same time that makes feature selection desirable, poses a severe challenge to feature selection algorithms, as most can not confront scalability issues and thus new methods should be devised [154,2,155]. Several more scalable feature selection algorithms have been developed during the last years, following online [156,100], or parallel and distributed strategies [157,73,72,71].…”

Supervised segmentation of MRI brain images using combination of multiple classifiers

“…Therefore, some researchers have proposed a combination of these methods together with Markov random field (MRF) algorithm for incorporation of contextual, textural and spectral information in segmentation process [7,8]. Ahmadvand et al [9] proposed a wavelet channel combining method for combining of different channels in multispectral MRI images and then, they used a novel combination of FCM algorithm with MRF method for robust multispectral MRI brain image segmentation. In these methods, MRF algorithm is a post-processing step for smoothing different segments and to remove the noise.…”