Predicting First-Year Engineering Student Dropout Using Data Mining: A Case Study at Salahaddin University-Erbil

High dropout rates among newly registered students in engineering programs significantly impact the academic plans and long-term strategies of many schools and universities. Student dropouts and low enrollment rates in engineering programs are major concerns for higher education institutions and governmental decision-makers in both developed and developing countries, however the reasons for these issues vary. This study utilizes a data mining approach to predict the features influencing dropout decisions. The method was applied to pre-enrollment and first-semester official data from the engineering college at Salahaddin University-Erbil. A dataset was built, preprocessed, modeled, and analyzed. The research yielded valuable results and suggested redesigning the application forms for engineering programs to further improve prediction rates and plan for student admittance who will complete the program successfully.