Analysis Of Undergraduate Student Performance Patterns and Profiling Potential Dropouts

Students' performance is a success component in greater education institutions. Predicting and analyzing the overall performance of the scholar is imperative to aid educators in figuring out weaknesses and bettering academic scores. However, accomplishing correct predictions is challenging due to the large quantity of instructional data. The important purpose in the back of this is the lack of research on exploring one-of-a-kind prediction strategies and key attributes that have an impact on the student's academic performance. This study found that the computer classification device mastery algorithm was most commonly used to predict student learning success. Four algorithms that have been used more frequently to predict learning about student success are ANN, naive boys, regression of logistics, SVM and decision tree. Meanwhile, the records used in these research papers mainly ranked students' success in the master's in two or three classes that are pass/fail; or fail/pass/excellent. The results of this research will give recommendations and contributions to policy makers and the higher education institutions in Somalia.