Research on Online Creativity Education for College Students Using Data Mining
Creativity education plays a very important role in the future development of students. This paper analyzed the effectiveness of two decision tree methods: C5.0 and classification and regression tree (CART) algorithms, using data mining, in predicting the performance of students enrolled in online education, discussed the data collection and processing methods, selected 12 features as input, and conducted an example analysis of the data set. It was found that the decision tree method achieved better performance and higher prediction precision in comparison with the nearest neighbor algorithm, and the accuracy and F1-measure value of the C5.0 algorithm was 92.59% and 92.53% respectively, which was better than the CART algorithm. The analysis of the decision tree demonstrated that the factors which had the greatest impact on students’ grades were the fifth and sixth chapter test scores, the number of questions answered correctly, etc. The experiment results verify the effectiveness of the decision tree method in predicting students’ performance in online creativity education, which can be further applied in practice.
Keywords: data mining, colleges, online education, creativity education, decision tree
K. Zhou, "Research on Online Creativity Education for College Students Using Data Mining", Engineering
Intelligent Systems, vol. 30 no. 6, pp. 447-452, 2022.