Prediction of optimum student performance factors using Genetic Algorithm

The Knowledge Discovery and Data Mining (KDDM) is a developing field of study contended  to be very extremely valuable  in finding  knowledge concealed  in huge  datasets. Higher Educational Institutions (HEIs) are slowly adopting KDDM processes. While literature shows that KDDM processes empower revelation of knowledge helpfull to further develop organizational performance, limitations encompassing them go against this contention. Examples of KDDM in the literature demonstrate the advantages in spite of the contradictions. From one perspective KDDM processess promise support to HEIs by providing great insight into the hidden knowledge that could not be easily understood. One region that noticeably stood apart as a significant test was the revelation obviously taking examples in instructive datasets related with context oriented data. The dataset related to students who had graduated between 2003 and 2014. The optimal CGPA (Cumulative Grade Point Average), time-to-degree, course difficulty level and course taking pattern were the attributes used to test the CRISP-DM model. When experiments were conducted using CRISP-DM process by applying Classification technique. When experiments were conducted using CRISP-DM process by applying Genetic Algorithm Classification technique. Classification produced course taking patterns that were partially linked to CGPA and time-to-degree. But optimum CGPA and time-to-degree could not be linked. Similarly the outcome of classification did not produce patterns with contextual information.