%0 Journal Article %A GENG Yinfeng %A ZHANG Xueying %A LI Fenglian %A HU Fengyun %A JIA Wenhui %A WANG Chao %T Optimized Extreme Learning Machine and Its Application in Classification of Stroke TCD Data %D 2020 %R 10.3778/j.issn.1002-8331.1901-0261 %J Computer Engineering and Applications %P 268-272 %V 56 %N 10 %X

In order to improve the efficiency and accuracy of Transcranial Doppler(TCD)data classification for stroke prediction, an Extreme Learning Machine(ELM) optimized by Bat Algorithm(BA) model is proposed. The element values of input weight matrix and threshold matrix in hidden layer are randomly set in the process of training the ELM model, which badly affect the classification performance of the network. To solve this problem, BA is used to optimize the parameters mentioned above. The proposed BA-ELM model is further used to classify the stroke patient TCD data in the experiment. Experimental results indicate that, the accuracy of the BA-ELM model is improved by 22.77% compared with the typical ELM model for classifying TCD data set, so the proposed BA-ELM model can effectively be used for stroke prediction.

%U http://cea.ceaj.org/EN/10.3778/j.issn.1002-8331.1901-0261