{"title":"Emotion Recognition System based on EEG signal: A Comparative Study of Different Features and Classifiers","authors":"M. K. Ahirwal, M. Kose","doi":"10.1109/ICCMC.2018.8488044","DOIUrl":null,"url":null,"abstract":"In this paper, emotion recognition system based on electroencephalogram (EEG) signals has been implemented. The new technology of emotion recognition using physiological signal is based on pattern recognition and classification problem that is well described in this paper. A very famous dataset of EEG signals for emotion recognition known as DEAP dataset is used and three categories of features, time domain features, frequency domain features and entropy based features has been extracted. Classification through support vector machine (SVM), artificial neural networks (ANN) and Naïve bayes (NB) has been done. Performance of system is observed by the parameters like, classification accuracy, precision and recall. After performance observation, it is found that ANN gives the best performance with all types of features. Highest classification accuracy achieved by ANN for said dataset, entropy based features and implementation is 93.75 percent.","PeriodicalId":6604,"journal":{"name":"2018 Second International Conference on Computing Methodologies and Communication (ICCMC)","volume":"28 1","pages":"472-476"},"PeriodicalIF":0.0000,"publicationDate":"2018-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"18","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 Second International Conference on Computing Methodologies and Communication (ICCMC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCMC.2018.8488044","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 18
Abstract
In this paper, emotion recognition system based on electroencephalogram (EEG) signals has been implemented. The new technology of emotion recognition using physiological signal is based on pattern recognition and classification problem that is well described in this paper. A very famous dataset of EEG signals for emotion recognition known as DEAP dataset is used and three categories of features, time domain features, frequency domain features and entropy based features has been extracted. Classification through support vector machine (SVM), artificial neural networks (ANN) and Naïve bayes (NB) has been done. Performance of system is observed by the parameters like, classification accuracy, precision and recall. After performance observation, it is found that ANN gives the best performance with all types of features. Highest classification accuracy achieved by ANN for said dataset, entropy based features and implementation is 93.75 percent.