132 Journal of Education and Vocational Research Vol. 3, No. 5, pp. 132-137, May 2012 (ISSN 2221-2590) Measuring Students’ Performance with Data Mining Jean Pierre Atanas Advanced University Program, the Petroleum Institute, Umm Al Naar, Abu Dhabi, United Arab Emirates jpierre@pi.ac.ae Abstract: Understanding the true reasons behind students’ failure, and bringing preventive measures to this issue at early stages are invaluable in the educational learning process. Preventing problems such as language deficiency or misclassification of the students in the appropriate academic levels is primordial for any educational institution. Many factors influence the learning process of the students, such as the demographic characteristics, educational background as well as language barrier. This work highlights the most preponderant factors affecting students’ advancement in the learning process and provides support to academic administrators. It uses some of state of the art classification and regression algorithms in the application domain of predicting students’ progress. Datasets were filtered and trained using predictive algorithms. It is shown that Science learning and English language skills are highly correlated. Datasets are not always suitable for data mining unless it is preprocessed and well adapted to the context being studied. A tool has been developed to preprocess the data provided that feeds into Weka Data Mining Software to profile students’ performance. Keywords: Education, correlation, measuring performance, prediction, students’ profiling 1. Introduction In recent years, researchers have started to apply machine-learning techniques to predict students’ performance and their learning styles (Pattanasri, Mukunoki, & Minoh, 2012).Tracking students’ performance is a preventive process and could be applied at early stages in order to identify poor performers and hence apply different learning styles. Remedial actions can be taken, on the spot, by tutors or administrators, to overcome this issue and provide additional help to minimize groups at risk. University placement and entrance exams administered to students are not always conclusive and predictive of the students’ performance as many factors can intervene throughout the academic year. Diagnosis of students’ performance is a dynamic process. It becomes more accurate as new curriculum information is entered during the academic year. Many machine learning algorithms have been used for the purpose of predicting performance(Kotsiantis, Pierrakeas, & Pintelas, preventing student dropout in distance learning systems using machine learning techniques, 2003);(Kotsiantis, Pierrakeas, & Pintelas, 2004);(Xenos, Pierrakeas, & Pintelas, 2002). None has been identified as the best algorithm for all cases (Mitchell, 2011). Accuracy of the outcomes depends on the quality of the data itself, the preprocessing phase, algorithms being used as well as the attribute being predicted. The following sections describe in brief the University Program and policies, the correlations between language skills and Science, an experiment results for all of the tested algorithms to profile students’ performance, a comparative study among learning algorithms and a conclusion. The AUP (i.e., Advanced University Program) is a preparatory program for one year, and is designed to help students develop the knowledge, study skills and work habits that are needed to prepare them to be successful at an excellent engineering university. While studying in this program, learners have the opportunity to earn university credit for courses in mathematics, chemistry and physics. Students who meet the requirements may apply these credits to their Bachelor degree, allowing them to proceed through this program in less time. English language and computing courses have been designed to help our students acquire the language, technological, and analytical skills required to meet entrance requirements fixed by the university and assist them in their future studies. The AUP English course is a one-year modular program that helps high school graduates prepare for the four- year engineering degree program through a set of four modules in the fall and one in the spring. The AUP English course also works to support AUP Math and Science to ensure student success in all subject areas. The