| 摘 要: 神经网络算法是深度学习研究的重点，遗传算法是一种自适应优化搜索算法，模拟退火算法是寻找最优 解的算法，本文主要分析了神经网络，遗传算法和模拟退火算法的特点和缺陷，研究BP神经网络和遗传模拟退火算法 相结合的技术，从发挥算法的优点基础上，提出了一个基于模拟退火遗传算法的BP神经网络模型，并应用于某观影俱 乐部，作为新电影上映预测和用户推荐，实验结果表明：该算法在收敛性和准确率上都有较好的效果。 |
| 关键词: BP神经网络；模拟退火；遗传算法；收敛 |
|
中图分类号: TP391
文献标识码: A
|
|
| A Predication Model of BP Neural Network Based on Genetic Simulated Annealing Algorithms |
|
JIANG Meiyun
|
( Department of Computer and Software, Nanjing Institute of Industry Technology, Nanjing 210023, China)
|
| Abstract: The Neural Network Algorithm is a focus of deep learning,Genetic Algorithm is to search adaptive optimization and Genetic Simulated Annealing Algorithm is to find the optimal solution.Analyzing the characteristics and defects of these three algorithms as well as studying the technology of combining BP neural network with genetic simulated annealing algorithm,this paper proposes a new BP neural network model based on genetic simulated annealing algorithm from getting the advantages of the algorithm,and the new algorithm has been used in a movie club,where the experimental results show that this algorithm has better effects on the convergence and accuracy. |
| Keywords: BP neural network;simulated annealing;Genetic Algorithm;convergence |
