Hamwira Yaacob

Systematic review of computational modeling of mood and emotion

In the recent years, more studies that aim to make computers understand, experience and respond to affects accordingly through computational models have been widely researched. Although many studies have defined and distinguished the words affect, mood and emotion, such terms are still used interchangeably. Thus, in this study, a systematic literature review was implemented to summarize and evaluate the current states of the arts on computational modeling of mood. From three online databases including IEEE Xplore, ScienceDirect and Springer Link, 825 scientific articles were extracted. Furthermore, through the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta Analyses) Statement, 9 articles were selected for the review. These resulting articles were reviewed based on several categories including the aim of the study, the study population, the measurement of mood, the basic emotion dimension, and proposed computational model, as well as evaluation. As a result, the systematic literature review has provided a good starting point in the study of the computational modeling of mood and emotion.

Classification of EEG signals using MLP based on categorical and dimensional perceptions of emotions

Emotions are frequently studied based on two approaches; categorical and dimensional. In this study, Multi-Layer Perceptron (MLP) was employed to classify four affective states as posited from these approaches. It was observed that emotional states viewed from the dimensional perspective are well discriminated using memory test. In addition to that, the dynamic for each of the four emotions were also presented, in which it was also indicated that an emotional state does not occur abruptly.

Two dimensional affective state distribution of the brain under emotion stimuli

Emotions are ambiguous. Many techniques have been employed to perform emotion prediction and to understand emotional elicitations. Brain signals measured using electroencephalogram (EEG) are also used in studies about emotions. Using KDE as feature extraction technique and MLP for performing supervised learning on the brain signals. It has shown that all channels in EEG can capture emotional experience. In addition it was also indicated that emotions are dynamic as represented by the level of valence and the intensity of arousal. Such findings are useful in biomedical studies, especially in dealing with emotional disorders which can results in using a two-channel EEG device for neurofeedback applications.

Statistical Approach for a Complex Emotion Recognition Based on EEG Features

This paper presents electroencephalogram (EEG) signals and normal distribution technique to recognize the complex emotion. In the recent years, there has been a trend towards recognizing human emotions, however not many researcher aware that human can recognize more than one emotion at one time. Thus, in this study, normal distribution is utilized to recognize the expected emotion. The feature extraction and classification were obtained using a Mel-frequency cepstral coefficients (MFCC) and multilayer perceptron (MLP). The correlation between human emotion and mood is also the essential point, since the mood can affected to the human emotion. The results show that the human emotions is strongly influenced by his initial mood.

CMAC-based Computational Model of Affects (CCMA) for profiling emotion from EEG signals

Several studies have been performed to profile emotions using EEG signals through affective computing approach. It includes data acquisition, signal pre-processing, feature extraction and classification. Different combinations of feature extraction and classification techniques have been proposed. However, the results are subjective. Very few studies include subject-independent classification. In this paper, a new profiling model, known as CMAC-based Computational Model of Affects (CCMA), is proposed), CMAC is presumed to be a reasonable model for processing EEG signals with its innate capabilities to solve non-linear problems through self-organization feature mapping (SOFM). Features that are extracted using CCMA are trained using Evolving Fuzzy Neural Network (EFuNN) as the classifier. For comparison, classification of emotions using features that are derived from power spectral density (PSD) was also performed. The results shows that the performance of using CCMA for profiling emotions outperforms the performance of classifying emotions from PSD features.

A Cognitive-Affective Measurement Model Based on the 12-Point Affective Circumplex

In the existing studies, the quantification of human affect from brain signals is not precise because it is merely rely on some approximations of the models from different affective modalities rather than the neurophysiology of emotions. Therefore, the objective of this study is to investigate the cognitive-affective model for quantifying emotions based on the brain activities through electroencephalogram (EEG). For that purpose, the recalibrated Speech Affective Space Model (rSASM) and the 12-Pont Affective Circumplex (12-PAC) were compared. Moreover, Kernel Density Estimation (KDE) and Mel-Frequency Cepstral Coefficients (MFCC) were used for feature extractions and Multi-Layer Perceptron (MLP) neural network was employed as the classifier. The results show that the MFCC-12PAC cognitive-affective model is the best model for all subjects. Furthermore, the results indicate that emotions are unique between participants and consistent throughout performing executive function tasks. Therefore, our empirical work has provided evidences that 12-PAC model may be adapted to improve the quantification of human affects from the brain signals. The analysis may be later expanded for the construction of an automated tool for the understanding of childrens emotion during intervention sessions with psychologists.

Extracting Features Using Computational Cerebellar Model for Emotion Classification

Several feature extraction techniques have been employed to extract features from EEG signals for classifying emotions. Such techniques are not constructed based on the understanding of EEG and brain functions, neither inspired by the understanding of emotional dynamics. Hence, the features are difficult to be interpreted and yield low classification performance. In this study, a new feature extraction technique using Cerebellar Model Articulation Controller (CMAC) is proposed. The features are extracted from the weights of data-driven self-organizing feature map that are adjusted during training to optimize the error obtained from the desired output and the calculated output. Multi-Layer Perceptron (MLP) classifier is then employed to perform classification on fear, happiness, sadness and calm emotions. Experimental results show that the average accuracy of classifying emotions from EEG signals captured on 12 children aged between 4 to 6 years old ranging from 84.18% to 89.29%. In addition, classification performance for features derived from other techniques such as Power Spectrum Density (PSD), Kernel Density Estimation (KDE) and Mel-Frequency Cepstral Coefficients (MFCC) are also presented as a standard benchmark for comparison purpose. It is observed that the proposed approach is able to yield accuracy of 33.77% to 55% as compared to the respective comparison features. The experimental results indicated that the proposed approach has potential for comparative emotion recognition accuracy when coupled with MLP.

Affective State Classification Through CMAC-Based Model of Affects (CCMA) Using SVM

A number of computational models have been proposed to perform emotion profiling through affective state classification using EEG signals. However, such models do not include both temporal and spatial dynamic of the signals. It is also observed that the performance of classifying emotion using the existing models produce high classification accuracy on one subject, but not on different subjects. Thus, in this paper CMAC-based Computational Model of Affects (CCMA) is proposed as feature extraction for the classification task. CCMA keeps the temporal and spatial dynamics of EEG signals to produce better classification performance. Using Support Vector Machine (SVM) as classifier, the features produce higher classification accuracy for heterogeneous test.

Revolving Traditional EEG Device into Mobile Architecture

This Electroencephalogram (EEG) has been widely used to capture brain signals for affective recognition through computational modelling of affective states based on valence, arousal and dominance. Not until recently wireless EEG machines are introduced to record brain signals during physical activities and high range of mobility. Nevertheless, such devices are expensive. Therefore, the aim of this study is to design a mobile EEG systems using an existing stationary EEG amplifier, EXG Brainmarker. The main component for the proposed mobile EEG system are the electrodes cap connected to EXG Brainmarker which will be powered up using portable rechargeable battery, Surface tablet to record and store EEG signals, Portable Wi-Fi device as a medium for remotely connection to monitor the Surface tablet. The proposed EEG mobile system was tested in a Simulation Centre at Akademi Laut Malaysia (ALAM). The data was obtained from three marine pilot who were given tasks to navigate ships in the simulated scenes. Findings and observations from the experiment are reported in this paper. It was also observed that there are several common movements took place when maneuvering the ship such as lift up their hand, turn their head to left and right and a few steps walking.

A Hasty Approach to ECG Person Identification

Using electrocardiogram (ECG) to extract identity, mood and behavioral information of individuals is a hot topic in biometric for the last 15 years. In an ECG signal, the region identified as the QRS complex is primarily used for classification of individuals. In this paper, we study an accepted method for identification where feature points are extracted from selecting random points within the QRS region and using the multilayer perceptron (MLP) method for classification. In our experiments, feature points are varied and processing time are measured to study the speed in processing feature points for identification. Our results shows accuracy performance cost and gains and the performance with respect to the number of feature points. Additionally, a different method in using 3-point of QRS complex that can provide best accuracy and time performance is presented. Our method though compromises accuracy proves to give faster results and may be usable for future applications in IoT.