Open AccessApplication Research of BP Neural Network Optimization Based on Firefly Algorithm
Author(s) -
Kun Wang,
Shixin Li,
Hai Zhang
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1802/3/032040
Subject(s) - firefly algorithm , artificial neural network , firefly protocol , computer science , convergence (economics) , algorithm , process (computing) , stability (learning theory) , optimization algorithm , throughput , data mining , artificial intelligence , machine learning , mathematical optimization , mathematics , wireless , particle swarm optimization , zoology , telecommunications , economics , biology , economic growth , operating system
Aiming at the problems of low prediction accuracy and premature convergence of traditional BP network prediction models, firefly algorithm FA is introduced into BP neural network model to optimize the optimization process of neural network weights and thresholds. This paper presents a BP neural network prediction model based on firefly algorithm. Taking Fujian port throughput data as an example, the application verification and comparative analysis of the improved prediction algorithm are carried out. The experimental results show that the improved FA-BP prediction algorithm has better performance in prediction accuracy and stability.