Predicting Academic Success in Higher Education Using Machine Learning Techniques and Classification Algorithms to Support Data-Driven Counseling
Author : S Swetha and B Asha
Abstract :
Predicting student academic performance has become an important focus in higher education as institutions aim to improve student retention and success rates. Traditional evaluation methods often rely only on examination results or manual assessment, which may not effectively identify students who require academic support at an early stage. To address this challenge, a machine learning–based academic success prediction system is proposed to analyze various student-related factors and provide data-driven insights for counseling and academic planning. The proposed system collects and processes data related to student demographics, attendance, assignment scores, study behavior, and previous academic records. By applying classification algorithms such as Decision Trees, Random Forest, and Logistic Regression, the system predicts the likelihood of a student achieving academic success or facing academic risk. These predictions help educators and counselors identify students who may need additional guidance, mentoring, or learning support. Experimental evaluation shows that machine learning classification models can accurately predict academic outcomes and significantly assist educators in making informed decisions regarding student support and intervention programs. The system contributes to improving academic performance, reducing dropout rates, and enhancing overall educational quality through intelligent data analysis.
Keywords :
Academic Success Prediction, Machine Learning in Education, Classification Algorithms, Educational Data Mining, Data-Driven Counseling, Student Performance Analytics.
