Abstract: Academic Analytics is extracting hidden patterns from educational databases. The main goal of this area is to extract hidden patterns from student academic performances and behaviors. One of the main topics in academic analytics is to study the academic performance of freshman students. Students enrolled in first year are the most vulnerable to low student retention in higher education institution. Research studies from different Higher Educational Institutions already indicated that early identification of students with academic difficulty is very crucial in the development of intervention programs. As such, early identification of potential leavers and successful intervention program(s) are the keys for improving student retention. The study will utilize the available enrollment and admission data. Feature selection technique will be used to determine significant attributes. The study aims to produce predictive and cluster model in which can early identify students who are in need of academic help and program interventions. The extracted predictive and cluster models will be evaluated using confusion matrix and be integrated in the decision support application.