Bridging The Gap In Educational Outcomes: Systematic Review Of Prediction Models Tailored For Diverse Disabilities In Higher Education
Year: 2024
Type: review
Abstract: Education is the primary factor of the socio-economic development of the community, yet disabled students often find their unique barriers hindering the progress of their academic and behavioral development.An overview of various prediction models in the high education sector that accommodate the needs of the students with different types of disabilities is presented in this paper as a part of problem-solving activities.This review affirms traditional frameworks that have a restricted scope, either regarding some particular or disabled populations or general student populations, which are inappropriate for disabilities' polymorphous structure that demands a holistic solution.We delve into cutting-edge data mining and machine learning methodologies, such as Decision Trees, Neural Networks, Random Forests, and Support Vector Machines, examining their efficacy in the recognition of academic and behavioral tendencies for students with disabilities.The analysis of the associated advanced methods like dimensionality reduction, feature selection, and the development of hybrid models are discussed concerning the good effects on predictive accuracy as well as mitigating the problems of overfitting and the 'curse of dimensions.'The vital points for explaining our success rates in predictions are the psychological, demographic and environmental aspects of the students, researchers will concentrate more on the holistic needs of the students.At the same time, the review of the trends in the embracement of machine learning with self-regulated learning strategies and adaptive educational technologies is made with the focus on ethics, data privacy, and scalability.New useful information from recent world studies is convinced to show the potentials, drawbacks, and parts that need more efforts in the future.The discrepancy between dynamic, inclusive, and context-aware models is too sensitive to be left unnoticed thus our results indicate that time has come to adopt such models that will predict not only results but also guide interventions for improving academic and behavioral success.By bridging the gap in educational outcomes for students with disabilities, the study will contribute to tangible and actionable recommendations for educators, researchers as well as decision-makers for an equitable and excellent higher education.
Author: Kumar Pradyot Dubey
Language: English
Subfield: Computer Science Applications
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4405868893