A novel hybrid optimization algorithm based on improved ACO and FNN

作者：Jiang Dao^*

来源：International Journal of Hybrid Information Technology, 2016, 9(4): 173-182.

DOI：10.14257/ijhit.2016.9.4.16

• Amount of information
• Decision variables
• Hybrid optimization
• Hybrid optimization algorithm
• Improved ant colony optimization
• Parameters optimization
• Selfadaptive
• UCI data

Due to the insufficient of the fuzzy neural network in solving complex problems, an improved ant colony optimization(ACO) algorithm is introduced into the fuzzy neural network in order to propose a novel hybrid optimization(APEACOFNN) algorithm in this paper. In the APEACOFNN algorithm, the self-adaptive pheromone evaporation factor strategy is used to dynamically adjust the pheromone evaporation factor on searching route in order to gradually lessen the amount of information between the optimal path and the worst path, and realize the full searching optimization for decision variable space. Then an improved ACO(APEACO) algorithm is obtained. Aiming at the parameters optimization problem of fuzzy neural network, the proposed APEACO algorithm is used to comprehensively optimize and select the parameters of fuzzy neural network in order to propose a novel hybrid optimization (APEACOFNN) algorithm. Finally, in order to test the effectiveness of the APEACOFNN algorithm, five UCI data sets are selected. The experimental results show that the proposed APEACOFNN algorithm takes the faster approximation objectives and higher solving accuracy.

• 出版日期2016