当前位置: X-MOL 学术Int. J. Commun. Syst. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A multi‐agent system based for solving high‐dimensional optimization problems: A case study on email spam detection
International Journal of Communication Systems ( IF 2.1 ) Pub Date : 2020-11-09 , DOI: 10.1002/dac.4670
Hekmat Mohammadzadeh 1 , Farhad Soleimanian Gharehchopogh 1
Affiliation  

There exist numerous high‐dimensional problems in the real world which cannot be solved through the common traditional methods. The metaheuristic algorithms have been developed as successful techniques for solving a variety of complex and difficult optimization problems. Notwithstanding their advantages, these algorithms may turn out to have weak points such as lower population diversity and lower convergence rate when facing complex high‐dimensional problems. An appropriate approach to solve such problems is to apply multi‐agent systems (MASs) along with the metaheuristic algorithms. The present paper proposes a new approach based on the MASs and the concept of agent, which is named MAS as Metaheuristic (MAMH) method. In the proposed method, several basic and powerful metaheuristic algorithms are considered as separate agents, each of which sought to achieve its own goals while competing and cooperating with others to achieve the common goals. Altogether, the proposed method was tested on 32 complex benchmark functions, the results of which indicated the effectiveness and powerfulness of the proposed method for solving high‐dimensional optimization problems. In addition, in this paper, the binary version of the proposed method, called Binary MAMH (BMAMH), was implemented on the email spam detection. According to the results, the proposed method exhibited a higher degree of precision in the detection of spam emails compared to other metaheuristic algorithms and methods.

中文翻译:

解决高维优化问题的多代理系统:以电子邮件垃圾邮件检测为例

现实世界中存在着许多高维度的问题,这些问题无法通过传统的传统方法来解决。元启发式算法已被开发为解决各种复杂和困难的优化问题的成功技术。尽管它们具有优势,但是当面对复杂的高维问题时,这些算法可能会发现诸如弱点,如较低的种群多样性和较低的收敛速度。解决此类问题的合适方法是将多代理系统(MAS)与元启发式算法一起使用。本文提出了一种基于MAS和智能体概念的新方法,即MAS元启发式(MAMH)方法。在提出的方法中,几种基本且功能强大的元启发式算法被视为单独的主体,每个国家都力求实现自己的目标,同时与他人竞争和合作以实现共同的目标。总共,该方法在32个复杂的基准函数上进行了测试,结果表明了该方法解决高维优化问题的有效性和强大性。另外,在本文中,该方法的二进制版本称为Binary MAMH(BMAMH),是在电子邮件垃圾邮件检测上实现的。根据结果​​,与其他元启发式算法和方法相比,该方法在检测垃圾邮件中具有更高的精确度。结果表明,该方法有效解决了高维优化问题。另外,在本文中,该方法的二进制版本称为Binary MAMH(BMAMH),是在电子邮件垃圾邮件检测上实现的。根据结果​​,与其他元启发式算法和方法相比,该方法在检测垃圾邮件中具有更高的精确度。结果表明,该方法有效解决了高维优化问题。另外,在本文中,该方法的二进制版本称为Binary MAMH(BMAMH),是在电子邮件垃圾邮件检测上实现的。根据结果​​,与其他元启发式算法和方法相比,该方法在检测垃圾邮件中具有更高的精确度。
更新日期:2021-01-04
down
wechat
bug