摘要
当前高校教育改革逐渐深入,如何提高教学质量已经成为当前高校教学的工作重点,而教学评价是提高教学质量的关键指标。因此,合理、科学的教学评价方法显得尤为重要。目前神经网络被广泛应用于高校教学评价,但是神经网络具有样本需求多、计算量大及容易陷入局部最优解的缺陷。针对这些问题,提出变量预测模型分类(Variable Predictive Model Based Class Discriminate,VPMCD)方法的全方位高校教师教学质量评价方法,并将其与BP神经网络进行仿真对比分析。结果表明,相对神经网络,VPMCD方法能有效提高评价准确率。
The reform of university education is gradually penetrating nowadays.And the key to university teaching is how to improve the quality of teaching,the essential index of which is teaching evaluation.Therefore,a reasonable and scientific teaching evaluation method seems to be very important.Neural network is widely applied in the area of university teaching evaluation.However,disadvantages exist.A big number of samples and massive calculation are needed in neural network,and local optimal optimization easily occurs.To solve these problems,variable predictive model based class discriminate(VPMCD),an evaluation method based on university teaching quilty,is proposed,followed with a simulation comparison analysis with BP neural network.The results show that VPMCD can effectively improve the calculation efficiency and prediction accuracy.
作者
彭延峰
刘燕飞
何宽芳
PENG Yan-feng;LIU Yan-fei;HE Kuan-fang(Hunan Provincial Key Laboratory of Health Maintenance for Mechanical Equipment,Hunan University of Science and Technology,Xiangtan,Hunan 411201,China;Engineering Research Center of Advanced Mining Equipment of Ministry of Education,Hunan University of Science and Technology,Xiangtan,Hunan 411201,China;School of Mechanical Engineering,Hunan University of Science and Technology,Xiangtan,Hunan 411201,China)
出处
《教育教学论坛》
2019年第35期67-69,共3页
Education And Teaching Forum
基金
国家重点研发项目子课题(2018YFF0212902)
国家自然科学基金项目(51805161)
湖南省自然科学基金青年项目(2018JJ3187)
2015年湖南省普通高等学校教学改革研究项目(湘教通[2015]291号)
