Abstract: Due to the existence of several cognitive factors which are difficult to quantify in the cognitive ability evaluation, this paper presents the gray partial correlation evaluation to build cognitive student model. Firstly, the relationship of mutual influences in factors is discussed to get weight values by calculating partial correlation coefficient. Secondly, the comprehensive evaluation model is built based on the definition of the gray class and the whitenization weight function used for getting gray evaluation weights. Finally the model is verified with real case, which shows that the model is beneficial to evaluate the level of students’ knowledge and provides guidance for learning, simultaneously gives a reliable conclusion.

Key words: agile learning, cognitive ability, cognitive student model, gray partial correlation evaluation method

摘要： 针对认知能力评价这类因素较多，且难以进行定量描述的问题，建立认知能力因素评价体系，利用灰色偏相关评价法建立认知型学生模型。首先分析各认知能力因素的相互影响关系，计算偏相关系数得到各认知能力因素的权重值。然后确定评价灰类，建立白化权函数，得到各认知能力因素关于各评价灰类的灰色评价权值，从而建立认知型学生综合评价模型。最后通过实例对方法模型进行验证。实例表明，该方法模型能够对学生知识水平作出评价并引导学生进行学习，其结果具有一定的可靠性。

关键词: 敏捷学习, 认知能力, 认知型学生模型, 灰色偏相关评价法