基于GA-BP神经网络优化算法的篮球教学质量评价  

Evaluation of Basketball Teaching Quality Based on GA-BP Neural Network Optimization Algorithm

作　　者：张瑞全[1] ZHANG Rui-quan(Ministry of Sports,Chuzhou City Vocational College,Chuzhou 239000,Anhui,China)

机构地区：[1]滁州城市职业学院体育部,安徽滁州239000

出　　处：《内蒙古师范大学学报（自然科学汉文版）》2023年第2期169-174,共6页Journal of Inner Mongolia Normal University(Natural Science Edition)

基　　金：安徽省社科规划资助项目(AHSKY2021D80);滁州城市职业学院院级重点资助项目(2021bmzx02,2021bmzx03);安徽省质量工程课程思政建设研究资助项目(2020kcszyjxm180)。

摘　　要：篮球教学质量是反应学习效果与教学方法可行性的关键指标,对于提升优化篮球课程的教学方法具有重要意义。基于主成分分析法提取16个关键性评价指标,设定GA-BP神经网络期望误差为均方误差;基于自适应梯度下降法进行网络训练,为遗传算法的不同种群赋予不同交叉概率与变异概率,利用移民算子沟通实现种群进化以此改进遗传算法;基于改进遗传算法确定BP神经网络的初始权值与阈值,将16个篮球教学质量评价指标权重输入到GA-BP神经网络模型中获得教学质量评价结果。实验结果表明,该模型输出篮球教学质量评价结果的误差最低、时间开销最少,是评价篮球教学质量的可靠方法。The quality of basketball teaching directly reflects the learning effect and the feasibility of teaching methods, which is of great significance for improving and optimizing the teaching methods of basketball courses. Firstly, 16 key evaluation indicators were extracted based on principal component analysis, the expected error of the GA-BP neural network was set as the mean square error and the network training was conducted based on the adaptive gradient descent method. Secondly, different crossover probabilities and mutation probabilities were assigned to different populations of the genetic algorithm, and the population evolution was achieved by using communication of the immigrant operator to improve the genetic algorithm.Finally, the initial weights and thresholds of the BP neural network were determined based on the improved genetic algorithm, and the weights of 16 basketball teaching quality evaluation indicators were input into the GA-BP neural network model to obtain teaching quality evaluation results. The results showed that the model output the basketball teaching quality evaluation results with the lowest error and the least time cost, and was a reliable means to evaluate the basketball teaching quality.

关 键 词：篮球教学 遗传算法 BP神经网络 权值 评价指标 

分 类 号：TP29[自动化与计算机技术—检测技术与自动化装置]