Data Mining for Engineering Schools

the supervision of the academic performance of engineering students is vital during an early stage of their curricula. Indeed, their grades in specific core/major courses as well as their cumulative General Point Average (GPA) are decisive when pertaining to their ability/condition to pursue Masters’ studies or graduate from a five-year Bachelor-ofEngineering program. Furthermore, these compelling strict requirements not only significantly affect the attrition rates in engineering studies (on top of probation and suspension) but also decide of grant management, developing courseware, and scheduling of programs. In this paper, we present a study that has a twofold objective. First, it attempts at correlating the aforementioned issues with the engineering students’ performance in some key courses taken at early stages of their curricula, then, a predictive model is presented and refined in order to endow advisors and administrators with a powerful decision-making tool when tackling such highly important issues. Matlab Neural Networks Pattern Recognition tool as well as Classification and Regression Trees (CART) are fully deployed with important cross validation and testing. Simulation and prediction results demonstrated a high level of accuracy and offered efficient analysis and information pertinent to the management of engineering schools and programs in the frame of the aforementioned perspective. Keywords-component; Educational Data Mining; Classification and Regression Trees (CART); Relieff tool; Neural Networks; Prediction; Engineering Students’ Performance; Engineering Students’ Enrollment in Masters’ Studies.