摘要
本文在对目前学分制高校大学生学习水平测评方法进行分析的基础上，运用人工神经网络（ANN）理论中应用最为广泛的BP网络技术，构建了对大学生学习水平进行评价的非线性评价模型。实际模拟运算结果与现行的测评方法的结果基本近似，且有其独特的特点。与传统的统计分析模型相比具有更好的容错性、鲁棒性和自适应性。避免了人工确定各指标或各层次权重带来的主观性，使得不同科目的成绩具有一定的可比性。可为高校大学生各类学习标兵、奖学金、三好学生等的评比提供客观、可靠的依据。
Based on the artificial neural network principles, the problem of synthetic evaluation of college students studying levels is studied, a nonlinear BP neural network model is developed. The practical simulation results of BP accords with the present evaluation results, and shows obvious characteristics. The nonlinear BP neural network model has strong ability for fault-tolerance, robust and self-adaptation, which can be applied widely to evaluation of scholarship, study model, general student.
出处
《工程数学学报》
CSCD
北大核心
2005年第8期44-48,共5页
Chinese Journal of Engineering Mathematics
基金
辽宁省教委科研项目.
关键词
人工神经网络
BP网络
学习水平
学分制
artificial neural network
BP network
studying levels
credit system
