Estimation single output with a hybrid of ANFIS and MOPSO_HS

Adaptive Neuro-Fuzzy Inference System (ANFIS) has gained popularity in recent years due to its predictive capabilities. Proper adjustment of ANFIS parameters is an optimization problem but integrating it with traditional optimization techniques has led to challenges such as local minima and slow convergence, resulting in obstacles to its prediction. Additionally, some researchers focusing on incorporating single-objective optimization often face issues with reliability and stability in parameter adjustment. This study, focused on multi-objective optimization, presents an algorithm that integrates ANFIS with MOPSO_HS. The proposed model, compared and applied to three real-world datasets, has demonstrated robustness in prediction problems. A comparative analysis is conducted between the proposed integrated model and well-known integrated algorithms with 20 runs. For further comparison, the Wilcoxon signed-rank test is used to determine whether there is a statistically significant difference in performance. The experimental results indicate the algorithm's accuracy, stability, and reliability in solving integration problems, highlighting its superiority over alternative approaches.