DEVELOPMENT OF ASIAN INSTITUTE OF TECHNOLOGY AND EDUCATION STUDENT SUCCESS PREDICTION SYSTEM (AITESSPS) USING J48 DECISION TREE ALGORITHM

The developed system aimed to predict AITE students’ success in terms of degree completion using J48 decision tree algorithm. This could help the school in developing intervention program for those students who are likely to encounter problems in completing their degrees. The proposed project used CRISP-DM and incremental model methodologies. CRISP-DM breaks down the life cycle of data mining project into six phases: business understanding, data understanding, data preparation, modeling, evaluation, and deployment. Meanwhile, incremental model followed the following phases: analysis, design, coding, and testing. Data preparation and cleansing was done to identify the best data predictors of student success. Data set was then loaded to WEKA data mining tool. J48 decision tree algorithm was utilized to classify the predicted result of student performance. Test option 10-folds cross validation was used. This technique partitions the data set into a training set to train the model and a test set to evaluate it. J48 generate decision rules, which was used in the development of the system. Unified Modelling Language (UML) was used to give user a clear picture of what the system does. Software tools such as Microsoft Visual C# and MySQL as relational database management system was used to develop the system. After the development of the project, it was tested and evaluated based on ISO25010 standard for software product quality in terms of functionality, efficiency, compatibility, usability, reliability, security, maintainability, and portability. There were 10 respondents who evaluated the project. Based on the evaluation result, the evaluators agreed the system acceptable in terms of functionality and efficiency.