Abstract
A new enhanced grasshopper optimization algorithm (GOA) has been developed and successfully applied to feature selection. GOA, as a heuristic algorithm, is proposed by simulating the living habits of grasshoppers in nature. Although GOA has an excellent global optimization capability, it still faces the disadvantage of low efficiency of searching optimization due to its ease of falling into the local optimum. Hence, based on the original GOA, this study integrates new ideas to reduce the defects to obtain a better global optimization ability. Because of the continuous optimization problem, the features of pursuing the best possible individual of spiral motion have been considered. The spiral motion is integrated into the GOA exploitation search stage, which further expands the diversification and intensification trends' capacities and effectively balances the exploration and exploitation procedures. Intuitively speaking, GOA with spiral search method can find better solutions in the exploration movement process, which is more efficient than the original search method. In the experimental comparison, to verify the proposed SGOA's ability in dealing with global unconstrained and constrained optimization problems, we compared it with other 30 IEEE 2017 benchmark tasks in meta-heuristic algorithms. Then, it is adopted to optimize engineering design and feature selection problems. We can know that the proposed SGOA has a good optimization ability in practical application from the experimental results. Spiral motion mode can significantly improve the original GOA's exploitation and exploration ability, and the proposed SGOA is of great assistance in practical fields. More info about this paper can be found on the web services https://aliasgharheidari.com.
Highlights
Optimization can be an essential task for realizing the feasible and unfeasible solutions with regard to specific superiority in terms of a specifit metric [1]
The capability of the developed SGOA can judge its optimization ability from multiple aspects based on 30 benchmark functions
The results exhibited that the spiral motion mechanism plays a vital role in further strengthening the central tendency of grasshopper optimization algorithm (GOA) and alleviating the immaturity of convergence
Summary
Optimization can be an essential task for realizing the feasible and unfeasible solutions with regard to specific superiority in terms of a specifit metric [1]. Scholars have developed various meta-heuristic algorithms (MA) and improved them to solve practical problems. Many novel and traditional methods are appropriate for solving specific types of optimization scenarios [3,4,5,6,7,8,9,10,11,12]. MA can be a solution to many potential fields like support vector machines [13, 14], feature selection [8, 15,16,17,18,19], healthcare [20, 21], extreme learning machine (ELM) [22,23,24,25,26,27], bankruptcy prediction [28,29,30,31,32], engineering design [12, 33,34,35,36,37,38], monitoring [39,40,41,42], intelligent damage detection [43,44,45], control systems [46], machine learning tools [47,48,49], image handling [50,51,52,53], IoT [54], medical image recognition [55, 56], and image segmentation [57, 58]
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
