Heliyon, Journal Year: 2024, Volume and Issue: 10(12), P. e32541 - e32541

Published: June 1, 2024

Decision-makers have consistently developed a range of classification models, each possessing unique features within the domain intelligent models. These endeavors are all directed toward achieving highest levels accuracy. In recent developments, two notable methodologies—reliable modeling and jumping approaches—offer specific advantages in formulating cost functions been recognized for their role enhancing classifier Specifically, methodology is based on aligning learning process with discrete nature goal, while reliable integrates reliability factor into paradigm. However, innovative combination, leveraging both accuracy factors guiding processes, leads to creation high-performing classifier. This addresses research gap tackling challenges, which remains core focus present study. To evaluate performance proposed jumping-based environmental decision-making, we considered ten benchmark datasets spanning various application domains. The numerical results demonstrate that Reliable Jumping-based outperforms traditional classifiers across studied cases. As result, approach proves be viable effective alternative other methods applications.

Language: Английский