Browsing by Author "Hamwira Sakti Yaacob"

Mental fatigue is a brain state that causes energy levels to decline due to engagement in a prolonged period of cognitive tasks, such as online learning. It results attention deficits and poor academic performance by reducing focus and engagement during online learning sessions. It is essential to address mental fatigue early because it leads to permanent cognitive deficits and emotional impairments. Existing subjective assessments like surveys, interviews and self-report questionnaires are time-consuming and often reported as bias, since humans cannot accurately measure their cognitive performance. Although there are numerous machine learning (ML) approaches for mental fatigue detection using electroencephalogram (EEG) signals, there is a lack of study on mental fatigue detection among online learners. Therefore, this study aims to use EEG-derived emotional states metrics for mental fatigue detection by identifying existing EEG-based emotional metrics and mental fatigue detection techniques through literature review, then developing and evaluating machine learning models. Thus, this study employs Emotiv Performance Metrics (EPM) including engagement, excitement, interest, stress and relaxation. The mental fatigue detection consists of EPM metrics collection using an Emotiv Insight headset, and ML model development including Logistic Regression (LR), Support Vector Machine (SVM), Multilayer Perceptron (MLP) and Ensemble Learning model. A total of 10 students participated in the data collection session. However, data from only 8 participants were included in the mental fatigue detection. The two participants were excluded based on the Chalder fatigue Scale (CFS). One participant pre-CFS score is 4, which defines that he is already in a fatigue state. Other participants pre-CFS and post-CFS score was less than 4, that indicates the participants did not fatigue during the learning session. The findings of descriptive analysis, correlation analysis and statical analysis show the significance of EPM between the fatigue and non-fatigue sessions. The mental fatigue detection models show reliable and consistent performance. Notably, the Ensemble model exhibits the most consistency and reliability in the subject-dependent analysis with average accuracy 0.91, F1-score 0.91, ROC AUC score 0.96 and cross validation 0.87, where subject-independent analysis with accuracy 0.75, F1-score 0.76, ROC AUC 0.83 and cross-validation mean 0.75. This evident that the subject-dependent models are more consistent than the subject-independent model. The findings are also evident that EPM emotional states are crucial features in the mental fatigue detection for online learners. Furthermore, the findings of this research lead to proposes a real-time mental fatigue intervention framework for online learners, where mental fatigue detection is achieved using EPM emotional metrics and Ensemble learning. Future studies need to investigate the influence factors of emotional states and mental fatigue to evaluate the mental fatigue intervention framework. A comprehensive investigation is required for subject-independent and subject-dependent based mental fatigue detection and intervention techniques.

Mental health has become a global concern because of the increasing number of cases related to mental illnesses over the years which places heavy burdens on the global societal well-being and economic growths. In general, mental health is always associated with the states of emotion which can be measured based on several affective dimensions. Emotion plays an important role in various aspects of human daily lives. It can be captured and measured from various responses, such as self-assessment, automatic nervous systems (ANS) responses, facial expressions, voice, speech text, body language and posture and the brain states. Although emotion can be easily acted, it is hardly detected through other forms of emotional expressions. However, such fraudulent is more apparent through the neurophysiology manifestations. The studies of emotion through neurophysiology manifestations are mostly analyzed using various brain imaging devices. Nevertheless, such studies require complex neuroscientific and mystifying cognitive science expertise, which are also rare and expensive. Thus, it can be quite challenging to operate such devices on certain groups of subjects including children under 12 years old. Therefore, in this research, electroencephalography (EEG) is employed due to its simplicity and lower cost. EEG is also used to capture brain temporal dynamic up to millisecond precision, which may not be possible with other brain imaging techniques. Thus, several computational models have been developed to understand emotional states from EEG signals analysis. However, it has been observed that the existing models do not consider both the temporal and spatial dynamics of EEG signals. Hence, the results are inconsistent, especially for subject independent analysis. Therefore, a new emotion profiling technique that aims to provide better performance for understanding and distinguishing different emotional states from EEG signals is introduced in this research. The new model is constructed based on a computational model of cerebellar known as the Cerebellar Model Articulation Controller (CMAC), because it is imbued with similar qualities of self-organizing and non-linearity which are also observed in the neural processes as captured by EEG. A component of CMAC architecture is also utilized to capture spatial dynamic, as well as temporal dynamic of EEG input. Thus, the new model is termed as the CMAC-based Computational Models of Affects (CCMA). The research methodology consists of five phases including awareness of problem, design, development, evaluation and conclusion. With that, the objectives of this research are to design, develop and analyze the CMAC-based Computational Models of Affects (CCMA). Validation of CCMA and the benchmarking models are based on the classification accuracy for subject-dependent homogenous memory profiling, subject-dependent homogenous cross validation profiling, subject-dependent heterogeneous memory profiling, subject-dependent heterogeneous cross validation profiling and subject-independent profiling. The results show that the CCMA is comparable with other models for subject-dependent analysis. In addition to that, it outperforms the others in subject-independent case. In conclusion, it is rationalized that the CCMA is not only viable as a classification model but it is also envisaged to have several potentials for future works.