Prediction Research on the Scale of Middle School Students in Jilin Province Based on BP Neural Network %A 李晓筝 %J Hans Journal of Data Mining %P 99-106 %@ 2163-1468 %D 2023 %I Hans Publishing %R 10.12677/HJDM.2023.132010 %X 本研究基于2005~2019年吉林省统计年鉴中的社会经济指标，利用Bootstrap样本扩充技术对数据进行数据增广，尝试应用神经网络算法构建社会经济指标与中学在校生规模关系间的反演模型，并与多因素线性回归方法建立的模型精度进行比较，意图探究社会经济基本情况对中学在校生规模的影响并针对构建的预测模型进行验证。结果发现：1) 第三产业从业规模与在校生规模的相关性优于第一产业和第二产业。2) 预测模型中的因子数 ≥ 3时，任意三因子组合的预测精度 ≥ 0.94，认为三因素模型满足预测需求为最优状态。3) 基于神经网络的科学研究技术服务及地质勘查业、交通运输仓储及邮政业和教育从业规模对中学在校生规模的预测效果最好R2 = 0.94，RMSE = 3.98。最后根据研究发现的问题讨论形成的机制并提出具有针对性的建议。

Based on the social and economic indicators in Jilin Statistical Yearbook from 2005 to 2019, the Bootstrap sample expansion technique was used to augment the data, this study attempted to use neural network algorithm to construct the inversion model of the relationship between social and economic indicators and the size of middle school students, and compared the accuracy of the model with the multi-factor linear regression method. The purpose is to explore the influence of social and economic conditions on the size of middle school students and verify the prediction model. The results show that: 1) The correlation between the employment scale in the tertiary industry and the student scale is better than that in the primary industry and the secondary industry. 2) When the number of factors in the prediction model is greater than or equal to 3, the prediction accuracy of any three-factor combination is greater than or equal to 0.94, and it is considered that the three-factor model satisfies the prediction demand as the optimal state. 3) The scale of scientific research and technical service based on neural network, geological survey, transportation and storage, postal and educational employment has the best prediction effect on the scale of middle school students, R2 = 0.94, RMSE = 3.98. Finally, according to the problems found in the research, the mechanism of formation is discussed and the corresponding suggestions are put forward. %K 在校生规模，神经网络，社会经济指标，Student Scale %K Neural Network %K Socio-Economic Indicators %U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=63648