Improvement of fuzzy neural network using mine blast algorithm for classification of Malaysian Small Medium Enterprises based on strength

• Main Author: Hussain Talpur, Kashif
• Format: Thesis
• Language: English; English; English
• Published: 2015
• Subjects: QA76 Computer software
• Online Access: http://eprints.uthm.edu.my/1327/

Fuzzy Neural Networks (FNNs) with the integration of fuzzy logic, neural networks and optimization techniques have not only solved the issue of “black box” in Artificial Neural Networks (ANNs) but also have been effective in a wide variety of real-world applications. Despite of attracting researchers in recent years and outperforming other fuzzy inference systems, Adaptive Neuro-Fuzzy Inference System (ANFIS) still needs effective parameter training and rule-base optimization methods to perform efficiently when the number of inputs increase. Many researchers have trained ANFIS parameters using metaheuristic algorithms but very few have considered optimizing the ANFIS rule-base. Mine Blast Algorithm (MBA) which has been improved by Improved MBA (IMBA) can be further improved by modifying its exploitation phase. This research proposes Accelerated MBA (AMBA) to accelerate convergence of IMBA. The AMBA is then employed in proposed effective technique for optimizing ANFIS rule-base. The ANFIS optimized by AMBA is used employed to model classification of Malaysian small medium enterprises (SMEs) based on strength using non-financial factors. The performance of the proposed classification model is validated on SME dataset obtained from SME Corporation Malaysia, and also on real-world benchmark classification problems like Breast Cancer, Iris, and Glass. The performance of the ANFIS optimization by AMBA is compared with Genetic Algorithm (GA), Particle Swarm Optimization (PSO), MBA and Improved MBA (IMBA), respectively. The results show that the proposed method achieved better accuracy with optimized rule-set in less number of iterations.