论文部分内容阅读
作为一种单隐层前馈神经网络,极限学习机(Extreme Learning Machine:ELM)相比传统神经网络算法具有模型简单、泛化能力好、学习速度快等优点,在大规模基因芯片技术的应用中为基因表达数据的肿瘤诊断提供了新的途径,是交叉科学领域新的突破.针对极限学习机随机确定权值,以及其算法存在大量隐层的神经元个数导致算法性能不稳定、分类精度不理想等问题,采用基于优化理论中的Fibonacci序列对ELM隐层节点与偏置进行改进,提出了一种基于Fibonacci优化理论的ELM分类方法(F-ELM).将改进分类方法应用到Hepatitis和Bridges数据集上,实验结果表明,基于Fibonacci优化理论的ELM分类方法性能得到提升,并相对传统的SVM算法、BP和Bayes算法的分类精度较高.
As a single hidden layer feedforward neural network, Extreme Learning Machine (ELM) has the advantages of simple model, good generalization ability and fast learning speed compared with the traditional neural network algorithm. In the application of large-scale gene chip technology, Which provides a new way for tumor diagnosis of gene expression data and is a new breakthrough in the field of cross-science.Aiming at the limit learning machine randomly determining the weight and the number of hidden neurons whose algorithm has a large number of hidden layers, the performance of the algorithm is unstable, (F-ELM) based on Fibonacci optimization theory is proposed to improve hidden layer nodes and bias of ELM by using Fibonacci sequence based on optimization theory. The improved classification method is applied to Hepatitis And Bridges datasets. The experimental results show that the performance of ELM classification method based on Fibonacci optimization theory is improved, and the classification accuracy of BP and Bayes algorithm is higher than that of traditional SVM algorithm.