Abstract: To distinguish students with different categories by cognitive and behavioral expression so as to better improve teachers’ teaching quality, a data mining model based on cognitive and behavior calculating models is proposed. Cognitive and behavior calculating models are proposed based on various important cognitive and behavioral input parameters. Artificial neural network, Sensitivity Analysis（SA）, data mining, Classification and Regression Tree（C＆RT） are applied to classifying by six cognitive parameters and three behavioral parameters. Students are classified to three categories so as to better implement different teaching strategies according to different categories of students. Experiments results show that the behavioral parameters are far more important than cognitive parameters. Analysis results indicate the feasibility of proposed models for supporting teachers’ work in educational systems.

Key words: cognitive and behavior calculating models, data mining, teaching quality improvement, artificial neural network, sensitivity analysis, classification and regression tree

摘要： 为了根据认知及行为表现区分不同类别的学生以更好地提升教师教学质量，提出了基于认知行为计算模型的数据挖掘模型。基于各种重要的认知、行为输入参数，提出了认知、行为指数因子计算模型；依据所搜集的六个认知参数及三个行为参数，运用人工神经网络、灵敏度分析、数据挖掘及分类回归树算法对数据进行分类；将学生划分成三种不同的类别，从而更好地针对不同类别的学生实施不同的教学策略。实验结果表明，学生分类问题中，行为参数远比认知参数重要，分析结果表明了所提模型在教育系统教师工作支持领域的可行性。

关键词: 认知行为计算模型, 数据挖掘, 教学质量提升, 人工神经网络, 灵敏度分析, 分类回归树