针对高维数据中存在冗余以及极限学习机(ELM)存在随机给定权值导致算法性能不稳定等问题,将限制玻尔兹曼机(RBM)与ELM相结合提出了基于限制玻尔兹曼机优化的极限学习机算法(RBM-ELM).通过限制玻尔兹曼机对原始数据进行特征降维的同时,得到ELM输入层权值和隐含层偏置的优化参数.实验结果表明,相比较随机森林,逻辑回归,支持向量机和极限学习机四种机器学习算法,RBM-ELM算法能获得较高的分类精度. In order to solve the problem of redundancy in high dimensional data and the instability of algorithm performance due to the existence of random weights for ELM, this paper presents a new method based on the restriction of Boltzmann machine (RBM) and ELM (RBM-ELM) .The optimization parameters of the ELM input layer weight and the hidden layer bias are obtained by limiting the dimensionality reduction of the original data by the Boltzmann machine.The experimental results show that RBM- Compared with the four machine learning algorithms of random forest, logistic regression, support vector machine and extreme learning machine, the RBM-ELM algorithm can obtain higher classification accuracy.