A Novel Two-tier Classifier based on K-Nearest Neighbour and Neural Network Classifier for Emotion Recognition using EEG Signals
Abstract- Emotion has a significant part in cooperation and correspondence between individuals. Emotion can be communicated either verbally through emotional vocabulary, or by communicating non-verbal prompts, for example, sound of voice, outward appearances and motions. As of late, estimation of human emotions from Electroencephalogram (EEG) signals assumes a crucial part in evolving academic Brain Computer Interface (BCI) devices. In this article we present a novel approach to classify human emotions using a two-tier classifier which is a combination of both K-Nearest Neighbour (K-NN) and Neural Network (NN) classifier. The useful informations for classification is extracted from the input EEG signals after noise removal which is done by the process of pre-processing. The proposed work is implemented on MATLAB and the simulation results are presented.
Keywords- EEG, Emotion Recognition System, Feature Extraction, K-NN, NN, Two-tier Classifier.
I. INTRODUCTION
Language and nonverbal interactions are used to switch over information with each other [1]. Communication between humans and machines or computer agents has become supplementary frequent as well as to human-tohuman communication [2]. Interaction between the human and the machine computers are never again seen as just computational trappings [3]. To interact with the users and the detecting capacities to conjecture user’s characteristics the imperative and conventional high-tech systems promptly succeed in several socio emotional life perspectives [4]. A choice of life aspects are e-health, training, telemonitoring of elderly people and learning [5]. Necessitate and consequence of habitual emotion recognition in the day by day existence of individuals emotions play an essential responsibility of human computer interface applications [6]. Enchanting emotions into account in the computer software could compose such applications further comfortable for users [7]. Exceptionally emotion recognition from the text, speech, facial expression or gesture is useful for medical applications, especially for aged people [8].
Read More