Student Graduation Prediction at Ibrahimy University Using K-Nearest Neighbor (KNN) Algorithm

Student graduation is an important indicator of a university's success in delivering quality education. Ibrahimy University faces challenges in objectively and proactively predicting student graduation, as academic evaluation processes remain conventional and reactive. This study aims to build a student graduation prediction system using the K-Nearest Neighbor (KNN) algorithm based on academic data including GPA, credits, attendance, and number of failed courses. The dataset consists of 150 student records from Ibrahimy University, developed using the Knowledge Discovery in Database (KDD) framework. Data was split into 80% training and 20% testing with StandardScaler normalization. The optimal k value was searched from k=1 to k=15. Results show that k=1 achieved the highest accuracy of 96.67%. The system is deployed as an interactive web application using Streamlit, enabling non-technical users such as lecturers and academic administrators to monitor student graduation potential more effectively and data-driven.