Introducing a New Classification Method using a CombinedApproach of Machine Learning and Multi-Criteria DecisionMaking

Decision-making issues have become very complicated and it is no longer possible to easilyassume the independence of criteria. Therefore, the use of Analytical Hierarchy Process(AHP) as one of the widely used methods in calculating the weight of criteria, which one ofits basic assumptions is non-dependence between criteria, has faced a problem. Therefore, inorder to optimize the parameters of the problem and increase the classification accuracy, theParticle swarm optimization algorithm will be used. The current research is developmental interms of its purpose, and quantitative in terms of data analysis method and mathematicalmodeling. In this paper, for the first time, a new hierarchical algorithm based on relationsbetween features will be presented for classification. In fact, in this article, for the first time,by presenting an improved and new version of the particle optimization algorithm, which willhave intersection and mutation operators, the ability to explore and search in the standardoptimization algorithm will be strengthened. Then, by using this new optimization algorithmand taking advantage of feature clustering and selecting the final features using the nodecentrality criterion, a new feature selection method has been presented. The results ofcomparative studies on credit datasets with different dimensions showed the very goodcompetitiveness of the proposed method in comparison with known machine learningmethods. Multi-criteria decision-making methods have often been used for ranking, while lessattention has been paid to the very good ability of these methods in data classification.Network analysis process in combination with particle swarm optimization algorithm showsan efficient and appropriate method in the field of data classification.