Research on FOA Optimization Extreme Learning Machine Algorithm and Model Application
The input weight and unit offset of traditional extreme learning machine algorithm are determined randomly,which leads to the bad performance of the algorithm,hence,the fruit fly optimization algorithm is adopted for improvement.The flow of fruit fly optimization algorithm(FOA)is analyzed.Three optimization algorithms,e.g.,fruit fly optimization algorithm,genetic algorithm and particle swarm optimization algorithm,are compared.FOA has good optimization performance by solving Schaffer function.Then,FOA is used to optimize extreme learning machine,and an improved extreme learning machine algo-rithm is proposed.The improved extreme learning machine algorithm and extreme learning machine algorithm are applied to the competitiveness prediction of chemical enterprises.The results show that the accuracy of the improved extreme learning ma-chine algorithm for the competitiveness prediction of chemical enterprises is significantly higher than that of extreme learning machine algorithm,and the difference in running time is very small.The FOA optimized extreme learning machine model pro-vides a reference for solving other prediction problems.
fruit fly optimization algorithmextreme learning machineenterprise competitiveness
万方数据
维普
