Predictive Modeling of Student Graduations using Artificial Neural Networks for Enhanced Educational Management: A Case Study of Kabanjahe 1 State High School

Abstrak

This research delves into the predictive modeling of student graduations at Kabanjahe 1 State High School, employing artificial neural networks as a cutting-edge tool for educational management. The study is driven by the overarching goal of developing a robust predictive model that aligns with actual graduation numbers, providing valuable insights for strategic planning and resource allocation. The methodology integrates historical enrollment and graduation data with advanced machine learning techniques. Comprehensive data preprocessing, feature selection, and the meticulous design of the neural network architecture lay the foundation for accurate predictions. The model's training and evaluation, marked by quantitative metrics, sensitivity analysis, and temporal assessments, attest to its accuracy, generalizability, and adaptability. The results reveal a predictive model with commendable accuracy, offering precise forecasts of student graduations. Temporal analyses unveil patterns and trends, enhancing the understanding of the model's consistency over time. Sensitivity analysis provides insights into influential variables, contributing to a nuanced comprehension of the factors shaping graduation outcomes. The implications of the findings extend beyond Kabanjahe 1 State High School, serving as a benchmark for similar educational institutions grappling with the challenges of dynamic student populations. The successful integration of artificial neural networks into the educational management framework establishes a precedent for technological integration in academic decision-making.