A5103267038- Online Learning and Analytics
- Innovative Teaching and Learning Methods
- EEG and Brain-Computer Interfaces
- Gaze Tracking and Assistive Technology
- Intelligent Tutoring Systems and Adaptive Learning
- Emotion and Mood Recognition
- Text Readability and Simplification
- Software Engineering Techniques and Practices
- Natural Language Processing Techniques
Monash University
2024-2025
Indian Institute of Technology Bombay
2021-2024
IITB-Monash Research Academy
2024
Delhi Technological University
2020
This work aims to investigate the performance of Long Short-Term Memory (LSTM) Model for EEG-Based Emotion Recognition. For experimentation, we use publicly available DEAP dataset, which consists preprocessed EEG and physiological signals. Our limits itself study only signals have a scope developing an efficient headgear model real-time monitoring emotions. In this study, extract band power, frequency-domain feature, from compare classification accuracies Valence Arousal domain different...
This paper addresses the problem of EEG-based emotion recognition and classification investigates performance classifiers for subject-independent subject-dependent models separately. The results are compared with other also existing work in concerned domain as well. We perform experiments on publicly available DEAP dataset band power feature accuracies found pertaining to widely accepted Valence-Arousal Model. best were reported by LSTM model case 94.69% 93.13% valence arousal scales...
Temporality in Self-Regulated Learning (SRL) has two perspectives: one as a passage of time and the other an ordered sequence events. Each these conceptions is distinct requires independent considerations. Only single analytic method not sufficient adequately capturing both facets temporality. Yet, most research uses temporally-focused SRL research, those that use multiple methods do address aspects We propose CTAM4SRL, consolidated temporal which combines advanced data visualisation,...
In the past, unraveling learner interaction data in TELE was a challenge. However, advent of LA has helped uncovering latent information log to scaffold learning. This paper focuses on VeriSIM, TELE, teach software design diagrams. The learners' performance system is used categorize them into three groups, namely, "full scorers", "partial and "give uppers". Our analysis found that full scorers spend significantly higher duration per action than give-uppers an introductory challenge presented...