A Novel Binary Emperor Penguin Optimizer for Feature Selection Tasks

Abstract

Nowadays, due to the increase in information resources, the number of parameters and complexity of feature vectors increases. Optimization methods offer more practical solutions instead of exact solutions for the solution of this problem. The Emperor Penguin Optimizer (EPO) is one of the highest performing meta-heuristic algorithms of recent times that imposed the gathering behavior of emperor penguins. It shows the superiority of its performance over a wide range of optimization problems thanks to its equal chance to each penguin and its fast convergence features. Although traditional EPO overcomes the optimization problems in continuous search space, many problems today shift to the binary search space. Therefore, in this study, using the power of traditional EPO, binary EPO (BEPO) is presented for the effective solution of binary-nature problems. BEPO algorithm uses binary search space instead of searching solutions like conventional EPO algorithm in continuous search space. For this purpose, the sigmoidal functions are preferred in determining the emperor positions. In addition, the boundaries of the search space remain constant by choosing binary operators. BEPO's performance is evaluated over twenty-nine benchmarking functions. Statistical evaluations are made to reveal the superiority of the BEPO algorithm. In addition, the performance of the BEPO algorithm was evaluated for the binary feature selection problem. The experimental results reveal that the BEPO algorithm outperforms the existing binary meta-heuristic algorithms in both tasks.

A Closer Look at Precision Hard Turning of AISI4340: Multi-Objective Optimization for Simultaneous Low Surface Roughness and High Productivity

“…The most beneficial advantage of this algorithm is that it does not face problems with the convergence of parameters as long as the population size is appropriately increased [ 35 ]. Not only can MOEPCA solve continuous optimization problems, but it has also evolved to solve binary problems [ 36 ] and multi-objective problems [ 37 ]. This evolution has helped in solving many problems, including but not limited to those related to cloud service providers and complicated network problems [ 37 ].…”

Feature Selection Problem and Metaheuristics: A Systematic Literature Review about Its Formulation, Evaluation and Applications

“…Long et al [180] investigated numerical optimization and feature selection through a butterfly-balanced optimization algorithm. In similar veins, the studies in Takieldeen et al [122] and Kalra et al [123] introduced the dipper-throated optimization algorithm and a novel binary emperor penguin optimizer, respectively, both serving feature selection tasks. Further contributions came from Tubishat et al [50], who delved into dynamic generalized normal distribution optimization for feature selection, and Li et al [187] designed a two-stage hybrid feature selection algorithm with applications in Chinese medicine.…”

“…The relationship between the formulation of the objective function and the chosen metaheuristic offers a lens into the evolving research preferences and trends in feature selection. Figure 30 The majority of articles with a "weighted multi-objective" formulation predominantly employ both "direct binarization" [73,80,82,87,94,95,98,101,104,105,108,112,115,118,119,123,133,134,136,149,150,152,153,158,159,161,169,177] and "binarization with various approaches" [78,89,91,92,96,106,107,110,114,124,129,131,132,138,142,144,146,154,…”

“…"Wine" [23,29,37,39,42,47,54,57,[64][65][66]68,71,[77][78][79]82,84,86,87,[89][90][91][92][93][94][95]97,[102][103][104][106][107][108][109][110][111][112][118][119][120][123][124][125]127,129,130,132,[134][135][136][137]…”

BIN_MRFOA: İkili Optimizasyon İçin Yeni Bir Manta Vatozu Beslenme Optimizasyonu Algoritması

“…Kalra et al presented the binary Emperor Penguin Optimizer algorithm for efficient solution of binary nature problems, leveraging the power of the standard Emperor Penguin Optimizer. The performance of the algorithm is evaluated over twenty-nine benchmark functions and binary feature selection problem [19]. Chantar et al proposed an advanced binary grey wolf optimizer within a wrapper feature selection approach for solving Arabic text classification problems.…”