Closing the gap: exploring the untapped potential of machine learning in deaf students and hearing students’ academic performance

• Published in: International Journal of Advanced Technology and Engineering Exploration Vol. 10; no. 108; p. 1449
• Main Authors: Raji, N R, Kumar, R MathuSoothana S, Biji, C L
• Format: Journal Article
• Language: English
• Published: Bhopal Accent Social and Welfare Society 01.11.2023
• Subjects: Academic achievement; Data mining; Deafness; Deep learning; Education; Feedback; Hearing; Machine learning; Performance prediction; Students
• ISSN: 2394-5443; 2394-7454
• DOI: 10.19101/IJATEE.2023.10101685

Assessments and critical feedback play a crucial role in helping students not only master a skill but also apply it effectively. Educational data mining (EDM) and machine learning (ML) tools are aiding educators in tailoring teaching strategies to individual student needs. While predictive analytics are widely used for hearing students, there is a notable gap in research on deaf students. Assessing deaf students necessitates the expertise of trained specialists, and their feedback is particularly critical in assisting these students in skill mastery. Various strategies have been developed to analyze the academic performance of deaf children, but there is a lack of integration of data to create a model categorizing different methods of early classification based on student academic performance. As part of a broader effort to address challenges faced by students struggling with speech perception and language development, there is an opportunity to conduct a systematic study of early academic interventions for deaf students. Failure to address these issues can result in an increased risk of delays in social-emotional development. The findings from our review highlight several key aspects, including (i) ML and EDM-based applications for student performance analysis, (ii) factors influencing academic performance among deaf students, (iii) potential EDM methods useful for assessing deaf children, (iv) the absence of benchmark data and the need for interpretability in existing methods, (v) the necessity for ML approaches in predicting the performance of deaf students, and (vi) the anticipated major assessment trend in the future through deep learning models. Our findings have implications for various stakeholders in education, including teachers, students, administrators, and researchers.