Abstract
The pattern, path, and discrimination according to the data can automatically discovered by the mathematical models of machine learning (ML), and accordingly the outcomes are applied to project the prospects and/or cause decisions as stated by the brand-fresh, unseen data. The supervised learning (SL) makes whole solutions are identified while generating projects about the solution by gathering information based on the labeled data. The most popular SL method is random forests (RF) that is adaptable and may be used to solve both grouping and regression issues. The RF training procedure is lengthy, resource centralized, and prone to wrong group as a result of these and other drawbacks. In this context, a combination of particle swarm optimization (PSO) based on weighted RF is presented in order to improve the efficacy of the RF.