The Social Engineering Optimizer (SEO). (June 2018)

• Record Type: Journal Article
• Authors: Fathollahi-Fard, Amir Mohammad; Hajiaghaei-Keshteli, Mostafa; Tavakkoli-Moghaddam, Reza
• Is Part Of: Engineering applications of artificial intelligence. Volume 72(2018)
• Page Start: 267
• Page End: 293
• Publication Date: 2018-06
• Subjects: Social Engineering Optimizer (SEO) -- Meta-heuristics -- Single-solution -- Optimization techniques -- Engineering applications; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2018.04.009
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: Although several meta-heuristics have been developed in the last two decades, most of them are population-based, undergo many steps along with several parameters that make them hard to understand and code. In addition, there are same procedures in recent metaheuristics which make them similar. So, the researchers usually are confused to select a metaheuristic and cannot find any superiority or at least in any algorithms. Because of this, the researchers still use the old algorithms instead of the recent ones. Contrary to previous work, this paper aims to develop a simple, intelligent and new single-solution algorithm that has just four main steps and three simple parameters to tune. Social Engineering Optimizer (SEO) starts with two initial solutions divided into attacker and defender. The attacker obtains the rules of Social Engineering techniques to reach its desired goals. By these simple features, the algorithm does precisely both intensification and diversification phases. The basis of the algorithm depends on how an attacker attacks to a defender by four different associated techniques. Finally, the proposed SEO is applied to solve a set of benchmark functions, important engineering and multi-objective optimization problems. The result shows its superiority in comparison with other well-known and recent meta-heuristics. Highlights: A new single-solution meta-heuristic approach inspired by Social Engineering is developed. The proposed easy and intelligent method has four main steps and just three simple parameters to tune. The SEO is tested on the different benchmark functions, engineering and also multi-objective optimization problems. The algorithm gives the opportunity to a user to select the suitable technique among four ones. The results confirm the performance and effectiveness of SEO in practice.