ABSTRACT
Feature selection (FS) is a crucial area of cognitive computation that demands further studies. It has recently received a lot of attention from researchers working in machine learning and data mining. It is broadly employed in many different applications. Many enhanced strategies have been created for FS methods in cognitive computation to boost the performance of the methods. The goal of this paper is to present three adaptive versions of the capuchin search algorithm (CSA) that each features a better search ability than the parent CSA. These versions are used to select optimal feature subset based on a binary version of each adapted one and the k-Nearest Neighbor (k-NN) classifier. These versions were matured by applying several strategies, including automated control of inertia weight, acceleration coefficients, and other computational factors, to ameliorate search potency and convergence speed of CSA. In the velocity model of CSA, some growth computational functions, known as exponential, power, and S-shaped functions, were adopted to evolve three versions of CSA, referred to as exponential CSA (ECSA), power CSA (PCSA), and S-shaped CSA (SCSA), respectively. The results of the proposed FS methods on 24 benchmark datasets with different dimensions from various repositories were compared with other k-NN based FS methods from the literature. The results revealed that the proposed methods significantly outperformed the performance of CSA and other well-established FS methods in several relevant criteria. In particular, among the 24 datasets considered, the proposed binary ECSA, which yielded the best overall results among all other proposed versions, is able to excel the others in 18 datasets in terms of classification accuracy, 13 datasets in terms of specificity, 10 datasets in terms of sensitivity, and 14 datasets in terms of fitness values. Simply put, the results on 15, 9, and 5 datasets out of the 24 datasets studied showed that the performance levels of the binary ECSA, PCSA, and SCSA are over 90% in respect of specificity, sensitivity, and accuracy measures, respectively. The thorough results via different comparisons divulge the efficiency of the proposed methods in widening the classification accuracy compared to other methods, ensuring the ability of the proposed methods in exploring the feature space and selecting the most useful features for classification studies.
ABSTRACT
Feature Selection (FS) techniques extract the most recognizable features for improving the performance of classification methods for medical applications. In this paper, two intelligent wrapper FS approaches based on a new metaheuristic algorithm named the Snake Optimizer (SO) are introduced. The binary SO, called BSO, is built based on an S-shape transform function to handle the binary discrete values in the FS domain. To improve the exploration of the search space by BSO, three evolutionary crossover operators (i.e., one-point crossover, two-point crossover, and uniform crossover) are incorporated and controlled by a switch probability. The two newly developed FS algorithms, BSO and BSO-CV, are implemented and assessed on a real-world COVID-19 dataset and 23 disease benchmark datasets. According to the experimental results, the improved BSO-CV significantly outperformed the standard BSO in terms of accuracy and running time in 17 datasets. Furthermore, it shrinks the COVID-19 dataset's dimension by 89% as opposed to the BSO's 79%. Moreover, the adopted operator on BSO-CV improved the balance between exploitation and exploration capabilities in the standard BSO, particularly in searching and converging toward optimal solutions. The BSO-CV was compared against the most recent wrapper-based FS methods; namely, the hyperlearning binary dragonfly algorithm (HLBDA), the binary moth flame optimization with Lévy flight (LBMFO-V3), the coronavirus herd immunity optimizer with greedy crossover operator (CHIO-GC), as well as four filter methods with an accuracy of more than 90% in most benchmark datasets. These optimistic results reveal the great potential of BSO-CV in reliably searching the feature space.