University Prescribed Student Uniform Classification Machine Learning Modeling

Abstract

This study evaluates the performance of three machine learning models, Support Vector Machine (SVM), Neural Network (NN), and Random Forest (RF), in classifying student uniform compliance at ISAT U Miagao using image data. The classification focused on three categories: Not in School Uniform, Female Complete Uniform, and Male Complete Uniform. Model performance was assessed using standard classification metrics. Among the models tested, the Neural Network consistently outperformed the others across all categories, demonstrating its effectiveness in accurately identifying uniform compliance through image data. The SVM also produced strong and reliable results, indicating its viability as an alternative model for this task. In contrast, the Random Forest model showed relatively weaker performance, particularly in recognizing students not in uniform, which may limit its effectiveness in high-accuracy monitoring environments. Overall, the findings highlight the superior capability of deep learning, particularly neural networks, in handling image-based classification tasks. The strong performance of SVM also supports the use of kernel-based approaches for institutional compliance systems. Meanwhile, the lower performance of Random Forest suggests potential limitations in more nuanced visual classification scenarios. These insights support the integration of advanced machine learning models in real-world applications that require reliable and automated compliance monitoring.