Early Identification of Vulnerable Students with Machine Learning Algorithms

• DOI: 10.37394/23209.2025.22.16
• Publication Date: 2025-01-27T11:48:03Z
• Authors: Mohammad Hamza Awedh; Ahmed Mueen
• Keywords: Identification

Education is an important component in defining the overall development of a country. It also significant tool for achieving success life. One aspects influencing any educational institution's its students' academic achievement. In institutions, student dropout complex problem. Educational managers consider it vital to predict student's risk dropping out as soon possible. still needs be easier accurately advance. The major problems present research work include overfitting predictive model, variable relationships, insufficient feature extraction, and data pre-processing complexity. key goal this study improve achievement, decrease number dropouts, create support plans, constantly modify these plans based on ongoing progress monitoring. Specifically, aims identify at-risk students early using machine learning algorithms, allowing institutions take timely targeted interventions. Identifying their time with you will ensure that vulnerable get they need, help prevent rates from increasing, significantly benefit general performance. work, King Abdulaziz University database was used. Exploratory Data Analysis (EDA) heavenly understanding characteristics data, identifying anomalies, recognizing trends, directing further pre-treatment procedures. Genetic Algorithm-optimized Latent Dirichlet Allocation (GA-LDA) used extraction. We utilize canopy clustering Gaussian Flow Optimizer (GFO) accurate grouping. Finally, hybrid Logistic Regression-K-Nearest Neighbour (LR-KNN) technique classification. Accuracy, precision, recall, F1-score, sensitivity, specificity metrics were examine proposed model.