摘要
传统的大脑情感学习模型因其结构特点及其训练算法的局限性,在高维度数据分类问题上表现不佳,为提高高维度数据的分类准确性,提出一种基于模拟退火算法改进的大脑情感学习模型。通过改进网络结构,并采用模拟退火算法优化大脑情感学习模型的训练过程,改善其拟合能力和局部搜索能力,提高模型对于高维度数据分类问题的分类准确率。选取UCI数据集中常用于算法性能对比的几组数据集进行实验,实验结果表明,对于维度较高的数据集,该模型具有较好的分类效果。
The traditional brain emotion learning model is not good in high-dimensional data classification because of its struc⁃tural characteristics and limitations of training algorithms.To improve the classification accuracy of high-dimensional data,a brain emotion learning model based on simulated annealing algorithm is proposed.By improving the network structure and using simulated annealing algorithm to optimize the training process of brain emotional learning model,its fitting ability and local search ability are improved,and the classification accuracy of the model for high-dimensional data classification problems is enhanced.Experiments are carried out on several sets of datasets commonly used in UCI datasets for performance comparison of experiments.The experimen⁃tal results show that the model has better classification effect for datasets with higher dimensions.
作者
唐詹
潘建国
TANG Zhan;PANJianguo(College of Information,Mechanical and Electrical Engineering,Shanghai Normal University,Shanghai 201400)
出处
《计算机与数字工程》
2020年第12期3017-3021,共5页
Computer & Digital Engineering
关键词
大脑情感学习
高维度
数据分类
模拟退火算法
brain emotional learning
high dimensional
data classification
simulated annealing algorithm
