SVM-Based evaluation model for college laboratory learning
Xiaoling Tan1, Zefu Tan1, Juan Qu2, Guangwen Xi3
1School of Electronic and Information Engineering, Chongqing Three Gorges University, Chongqing 404100, China
2School of Mathematics and Statistics, Chongqing Three Gorges University, Chongqing 404100, China
3Network Centre, Chongqing Three Gorges University, Chongqing 404100, China
Evaluation for laboratory learning is based on different factors, while each factor is varied by individuals. Hence it is difficult to express the quantitative nonlinear functional relationship among the evaluation indexes. With limited sample, Support Vector Machine (SVM) could be generalized by compromising between model’s complexity and learning ability. That is its advantage on the evaluation of small sample, nonlinear and multi-indexes. It is a good try to apply Support Vector Machine (SVM) to laboratory learning evaluation. With Support Vector Machine (SVM), the relationship between the learning quality and evaluation indexes could be revealed. Experiments show that Support Vector Machine (SVM) model is with high prediction accuracy, faster speed and simple algorithm. It is suitable and more reasonable for laboratory learning evaluation.