International Journal of Computer Applications (0975 – 8887) Volume 146 – No.7, July 2016 17 Classification of Human Emotions using EEG Signals Pravin Kshirsagar Research Scholar, Department of Electronics Engineering, Rajiv Gandhi College of Engineering and Technology, Chandrapur (M.S.) India Sudhir Akojwar Senior Member IEEE, Professor, Department of Electronics Engineering, Rajiv Gandhi College of Engineering and Technology, Chandrapur (M.S.) India ABSTRACT In this paper we proposed new features based on wavelet transform for classification of human emotions (disgust, happy, surprise, fear and neutral). from electroencephalogram (EEG) signals.EEG signals are collected using 64 electrodes from twenty subjects and are placed over the entire scalp using International 10-10 system or international 10-20 system. The EEG signals are preprocessed using filtering methods to remove the noise. Feature extraction of the principle signal is done by using methods such as wavelet transform. The feature extracted signals are then classified using Neural Network (NN) and the neural system is trained and we get trained classifier according to the classification of the signals and the results are obtained. To test the signal the feature extracted signals are given directly to the trained classifier and results are obtained. Keywords Electroencephalogram (EEG); Neural Network (NN); Wavelet transform 1. INTRODUCTION The inner state of a person “Emotions” play a important or vital role in analyzing the state of mind [1].Emotion is one of the most essential features of humans. Without the capability of emotions processing, computers and robots are not communicates with human in natural way [2].Emotion is often interred twined with mood, temperament, personality, disposition, and motivation. Emotions are complex they are a state of feeling that results in physical and psychological changes that influence our behavior. In emotion assessment using EEG signals, the time duration of EEG signals under given, number of channels, emotional stimuli, frequency bands, nature of statistical feature extraction methods and features plays an significant Role. The traditional tools for the investigation of human Emotional status are based on the statistical analysis and Recording of physiological signals from the both central and autonomic nervous systems (CNS and ANS). In this, we are considered the signals like valance, arousal, dominance and liking. The emotional experiences can be subdivided into 2 dimensions called as valence (how negative or positive the experience feels) and arousal (how enervated or energized the experience feels).These 2 dimensions are depicted on a 2-D coordinate map [3]. This 2-dimensional map was theorized for capturing one essential component of emotion which is known as core affect. Emotions can also been described as a biologically given and a result for evolution because they provided good solutions to ancient and recurring problems that faced our ancestors. Feelings are best understood as a subjective representation of emotions, private to the individual experiences them. Moods have diffuse affective states that generally last for much longer durations than emotions and are also generally less intense than that of the emotions. Affect is an encompassing term, used to describe the topics of emotion, moods and feelings, together, even though it is commonly used interchangeably with emotion. In addition, relationships exist between emotions, like having positive or negative influences, with direct opposites existing .In addition, relationships exist between emotions, such as having positive or negative influences, with direct opposites existing. Parts of the brain activated for different Emotion sets Happy - The left side of the frontal lobe - the left prefrontal cortex. Sad - The right side of the frontal lobe –the right prefrontal cortex. Laughter and humor - Temporal lobe and hypothalamus Anger - Limbic center of the brain The small structure within the limbic system is termed as the amygdale. Fast eye moment sleep - It is the cerebral cortex. .