Combining Different Classifiers in Educational Data Mining

Educational data mining is a crucial application of machine learning.The KDD Cup 2010 Challenge is a supervised learning problem on educational data from computer-aided tutoring. The task is to learn a model from students’ historical behavior and then predict their future performance. This paper describes our solution to this problem. We use different classification algorithms, such as KNN, SVD and logistic regression for all the data to generate different results, and then combine these to obtainthe final result. It is shown that our resultsarecomparable to the top-ranked ones in leader board of KDD Cup 2010.