EEG CLASSIFICATION OF HAND MOVEMENT USING MACHINE LEARNING

Abstract

Brain computer interface can provide a communication pathway and control channel between brain and external devices. In this paper, we used global EEG dataset from UCI Machine Learning Repository (https://archive.ics.uci.edu/ml/datasets/ Planning +Relaxx) to classify the dataset associated with the left and right hand movement. Initially normalization is used to preprocess the dataset. This paper gives the result of deploying two classification algorithm random forest and Support Vector Machine (SVM) classifier to classify the dataset. Random forest and SVM got accuracy of 98.46% and 73.84% respectively. After preprocessing, the processed dataset was input into random forest and SVM classifier. Comparing with this two accuracy, the accuracy result of random forest classification algorithm is quite good and promised to be used in BCI context.