Licensure Examination performance is a growing concern of most of the educational institution because it is one of the determinants of quality education and validates high quality instruction. Educational institution is focused on monitoring and improving the Licensure Examination performance particularly in Teacher Education Institution (TEI). The study intends to offer a possible solution to most TEIs apprehensions regarding LET performance by providing the students of Teacher Education a student support service in the form of a personalized Learning Management System with performance prediction and recommendation capability. This can be developed through drawing data model using several data mining techniques and tools. Previous literature suggested using data mining to classify students, predict student performance, improve student retention, enhanced student achievement and assess complex students? behavior to name a few. This research project will provide the groundwork for the generation of a prediction model an innovative Student Support System with an Integration of Information Technology that would help the students by providing a Learning Management System (LMS) with personalized learning environment, performance prediction and recommendation engine that can be beneficial to board programs. This particular study covers the identification of the appropriate data mining algorithms to be used in modeling the PLMS. After an intensive investigation and literature review conducted, the research the result shows that ID3 and J48 is the best data mining algorithms is student performance prediction. The researcher will be using these algorithms in the development of the PLMS to integrate the prediction function of the prototype.