基于SOM神经网络的工程经济学教学质量评价模型研究  

Research on teaching quality evaluation model of engineering economics based on SOM neural network

作　　者：李可心 张斌 蒙彦宇 王淋 LI Kexin;ZHANG Bin;MENG Yanyu;WANG Lin(School of Civil Engineering and Transportation,Beihua University,Jilin 132000,China)

机构地区：[1]北华大学土木与交通学院,吉林吉林132000

出　　处：《现代电子技术》2023年第18期162-166,共5页Modern Electronics Technique

基　　金：2021年度北华大学教育教学改革项目:融合人工智能的土木工程人才培养模式探索(XJQN2021033);2023年度吉林省教育厅高等教育教学改革研究项目:“人工智能+X”背景下的高校工程管理专业复合型人才培养体系研究(JLJY202352071018);2020年度北华大学教育教学改革项目:人工智能与高等教育工程管理专业建设交叉融合探索与实践(J2020003);北华大学研究生教育教学改革研究与实践项目:基于线上线下混合式的《弹性力学及有限元》课程立体化教学改革创新性研究与实践(JG2021005)。

摘　　要：针对传统教学数据处理及分析受多种复杂因素干扰,存在主观性强、数据处理效率低、评估结果不准确等问题,提出一种基于自组织特征映射(SOM)神经网络构建工程经济学教学质量评价模型的方法。首先,基于课程属性及特点设计可用于工程经济学教学质量评价的评价指标及其评价内容;然后,重点介绍基于SOM神经网络构建工程经济学教学质量评价模型全过程,包括量化评价指标并将其作为标准样本输入网络模型,调整模型各项参数使其性能达到最优,基于获胜神经元及其他神经元拓扑结构进行质量评价分析。最后,基于建设工程经济学课堂采集的教学评价数据,利用提取的12个教学质量评价指标,从理论及实践双维度分析教学数据,验证所提模型的有效性。结果表明,基于SOM神经网络构建的工程经济学教学评价模型可有效地对教学质量做出客观准确的评价结果。该方法可为实现工程经济学高效、准确的教学评估和推动智能化教学评价体系的构建提供参考。As the traditional teaching data processing and analysis are interfered by many complicated factors,resulting in some problems such as strong subjectivity,low data processing efficiency,and inaccurate evaluation results,a method for constructing an engineering economics teaching quality evaluation model based on self organizing feature mapping(SOM)neural network is proposed.Based on the course attributes and characteristics,the evaluation index and its evaluation content are designed,which can be for the teaching quality evaluation of engineering economics.The entire process of constructing an engineering economics teaching quality evaluation model based on SOM neural network is emphasized,including quantifying evaluation indicators and inputting them as standard samples into the network model,adjusting various parameters of the model to achieve optimal performance,and conducting quality evaluation analysis based on the winning neuron and other neuron topology structures.Based on the teaching evaluation data collected in the classroom of"construction engineering economics",the extracted 12 teaching quality evaluation indicators are used to analyze the teaching data from both theoretical and practical dimensions,so as to verify the effectiveness of the proposed model.The results show that the engineering economics teaching evaluation model constructed based on SOM neural network can effectively give objective and accurate evaluation results for teaching quality.This method can provide reference for achieving efficient and accurate teaching evaluation in engineering economics and promoting the construction of an intelligent teaching evaluation system.

关 键 词：工程经济学课程 教学质量评价 自组织特征映射神经网络 教学数据参量 评价指标 聚类分析 

分 类 号：TN711-34[电子电信—电路与系统] TP183[自动化与计算机技术—控制理论与控制工程]