基于BP神经网络算法的判罚对足球比赛成绩影响的预测模型研究  

Prediction Model of the Impact of Refereeing Decisions Based on BP Neural Network Algorithm on Football Match Results

作　　者：朴柳美 贾洪淳[1] Piao Liumei;Jia Hongchun(School of Physical Education,Beihua University,Jilin 132013,China)

机构地区：[1]北华大学体育学院,吉林吉林132013

出　　处：《体育科技文献通报》2024年第9期40-43,共4页Bulletin of Sport Science & Technology

基　　金：吉林省中小学课后体育服务多元保障体系建设研究(项目编号:GH23708)。

摘　　要：本文基于某省近3届运动会足球比赛的判罚数据,采用BP神经网络算法构建预测模型,分析判罚对足球比赛成绩影响的关联效应。对数据进行预处理和特征提取,并输入到神经网络中进行训练,构建出较为理想的预测模型。结果表明:(1)在BP网络选用的隐含层节点数为6时,其训练集的预测相对均方差较小,预测未来比赛结果的准确率达到了90%以上,且召回率和F1分数等评价指标均表现良好。(2)判罚情况与足球比赛成绩呈显著正相关,是影响足球比赛成绩的重要因素。本文为判罚对足球比赛成绩影响的预测提供了新思路,同时为判罚对其他体育比赛成绩的分析和预测奠定了理论基础。This study constructs a prediction model based on BP neural network algorithm using the refereeing data from the last three provincial sports meets in football matches,analyzing the correlation effect of refereeing decisions on match results.By preprocessing the data and extracting features for input into the neural network for training,an effective prediction model is established.The results indicate that:(i)when the number of hidden layer nodes in the BP network is set to 6,the relative mean square error of the training set predictions is relatively small,achieving an accuracy rate of over 90%in predicting future match results,with evaluation metrics such as recall rate and F1 score performing well.(ii)The refereeing situation is significantly positively correlated with football match results,serving as an important factor influencing match results.This study provides a new perspective for predicting the impact of refereeing decisions on football match results and lays a theoretical foundation for analyzing and predicting the effects of refereeing in other sports competitions.

关 键 词：BP神经网络 判罚 足球比赛成绩 预测模型 

分 类 号：G843[文化科学—体育训练]