中文摘要：

利用决策树算法出色的数据分析能力和直观易懂的结果展示等特点,采用C4.5算法挖掘学生在线学习行为与学习效果的历史数据.为避免决策树＂过拟合＂问题,在已生成的决策树上采用PEP方法进行剪枝,并构建学习分析模型.最后,利用建立的分析模型对测试数据集进行评估,得到了较为理想的分类预测结果.学习分析模型的创建为科学、合理地评估学生在线学习行为提供了有效的方法和手段,同时也给教学设计和课件开发提供了参考性建议.

英文摘要：

Along with the continuous popularization of online learning,online learning,like a treasure behavior,is hidden in the Network platform.A reasonable mining technology was need to discover and use it.Based on the characteristics of the decision tree like algorithm good data analysis ability and intuitive result display,the C4.5algorithm was used to explore the historical data of students＇ online learning behavior and learning efficiency in this paper.In order to avoid the over fitting problem of the decision tree,the decision tree that has been generated using method of PEP was pruned,and construct learning analysis model.Finally,the test data was evaluated by using the established analytical model,and obtained the comparatively ideal classification forecast result.The establishment of the learning analysis model provides an effective method and means for the scientific and reasonable assessment of students＇ online learning behavior,but also provides a reference for teaching design and courseware development.