Mining Students’ Data to Predict and Evaluate Their Academic Performance ( Case study: Faculty of Science, University of Nyala )

Abstract

Recently, institutions of Sudanese higher education testify a technological development in field of computerization management of student information electronically. Many data mining techniques are proposed to extract the hidden knowledge from educational data. Therefore, the educational data mining make it possible to extract these educational data and find information that assistant supporting both teachers and students. Moreover, the extracted knowledge and predicting performance helps the institutions to improve their teaching methods and learning process. These improvements lead to enhance the performance of the students and educational outputs. The aim of this study to design model to predicting the student’s performance based on data mining techniques. The data mining tool used to evaluated performance of student’s called Weka. The data consists of socio-economic, demographic and academic information included six hundred undergraduate students with eleven attributes .Classification task method was used the classifier tree j48, to predict the final academic results and grades of students in first year, Apriori algorithm was also applied to find the association rule mining among all the attributes and the best rules were also displayed. The results showed that classification process succeeded in training set. Thus, the predicted instance is similar to the training set. This proves the suggested classification model. The algorithm efficiency and effectiveness of j48 algorithm in predicting the academic results, grades, was very good. Furthermore recommend Test other algorithms to design a predictive model ,other than j48 and Increase the size of student data to create another model and compare it with the designer model.