Beyond Grades: Leveraging AI to Model Holistic Student Success Using Behavioural and Emotional Data

ABSTRACT : The older academic evaluation systems emphasize and focus on cognitive success and standardized testing, without paying much attention to non-academic aspects of success, such as emotional intelligence, motivation, resilience, and engagement, which are the core competencies to succeed in the 21st century. In this paper, we are suggesting an artificial intelligence model that considers both behavioral and emotional data to assess student success in a more comprehensive manner. Utilizing a variety of digital touchpoints, such as learning management systems (LMS), facial emotion recognition, and sentiment analysis, the proposed research generates a multi-dimensional profile of student development. In integrating these data sources, this research promotes a model where educational assessment becomes aligned with the current objectives of pedagogy and well-being. Machine learning models trained on structured interviews and surveys and validated across multiple institutions in various emerging economies, the study adds a scalable and confirmed method that refutes the simplistic grade-based measures, considering important ethical, cultural, and methodological issues.