摘要
针对目前智能教学系统中,大部分对学生认知能力的评价过于单一,而且并没有对学生的学习数据进行预处理,导致学生认知能力的评价结果缺乏说服力。本文提出在使用试题重复答对率对学生的学习数据进行预处理的基础上,采用矢量记录法量化的学习行为数据,结合专家给出的认知能力的起评参数,运用模糊综合评价方法评价学生的认知能力,从而构建出认知学生模型。通过该方法构建的认知学生模型可以有效地控制学生的不真实的数据,从而更贴切地刻画学生认知能力,使评价更具有合理性。实验表明该方法可行,并取得了良好的效果。
Since most contemporary intelligent tutoring systems do not pretreat students' learning data before they assess their cognitive ability, the results are always not comprehensive and convincing enough. This paper proposes an alternative approach to building students' cognitive model which features the following advantages:(1) students' learning data would be pretreated based on their repetitive correctness rate in testing;(2) the vector recording method would be employed to quantify students' learning behavior data; and(3) initial assessment indexes set by experts would be combined with fuzzy comprehensive evaluation. This paper argues that this approach would help produce truer data, and make the assessment more reasonable, thus students' cognitive ability could be better understood. The result of the experiment proves that it verifies the feasibility of this approach and obtains good effects.
出处
《广州广播电视大学学报》
2014年第2期30-34,107,共5页
Journal of Guangzhou Open University
关键词
智能教学系统
认知能力
模糊综合评价
学生模型
intelligent tutoring system
cognitive ability
fuzzy comprehensive evaluation
student model
