Comparison of analysis results obtained from WEKA and Orange data mining software in educational data mining

In this study, WEKA and Orange programs, which are frequently used in data mining, are compared based on the classification methods that can be used in data mining in education. The results of the ABIDE exam conducted by the Ministry of National Education were used in the study. During the analysis process, k-nearest neighbor, random forest, support vector machine, naive bayes and artificial neural networks methods were used to estimate the data of students classified according to their Turkish course scores by using demographic and psycho-social variables. In the second phase of the research, the reliability and cross validity values of the measurement results obtained from the students who were classified with the classification algorithms determined within the scope of the research were examined. According to the results obtained within the scope of the research; It was observed that Orange in k-nearest neighbor and artificial neural networks algorithms and WEKA in support vector machine and naive bayes algorithm have higher correct classification rates. Random forest algorithm has been found to have higher correct classification rate in binary and quinary classification in WEKA and Orange package program, respectively. In addition, the artificial neural networks algorithm obtained the highest correct classification rate in the scores classified as binary and quinary.