Data Analysis And Prediction Based On K12 Education Morning Inspection System

With the rapid growth of China's economy,people's lives have reached a well-off level.At the same time,the physical health of adolescents has also attracted widespread attention.In addition,with the opening of the second-child policy,the youth population in China has increased dramatically,and the health problems of adolescents have become urgent.Because adolescents' physical function is not perfect and their body resistance is relatively weak,some infectious diseases are easily spread among adolescents.For these easily spread diseases,the best treatment is to prevent them in advance.Therefore,the prevention of infectious diseases is receiving more and more attention.In order to provide a theoretical basis for the prevention of some diseases in primary and secondary schools in the future,a model that combines seasonal time series with neural network is proposed in this paper to analyze and predict the morning inspection data of primary and middle school students in Hubei in the past two years.First,the morning inspection data of the primary and secondary schools was obtained in the past two years.Then,the data was divided into test sets and prediction sets through preliminary processing and a SARIMA model was established to analyze and predict the test set.Next,the RNN-LSTM nerves were proposed to train the test set data for obtaining the trained neural network model.Finally,the predicted data of the SARIMA model is input into the RNN-LSTM neural network to correct the data and obtain the optimal predicted value.The main tool of this system is pycharm software.SARIMA model and LSTM neural network modeling are realized and optimized to use Python language programming.And the analysis results are displayed by pycharm's visual graphics tool.A large number of experimental analyses show that the short-term prediction of the SARIMA model is accurate but the long-term prediction ability is insufficient.However,long-term accurate prediction could be achieved by correction of the neural network.It can be seen that the combination SARIMA with LSTM can improve the accuracy and effectiveness of the prediction.The risk assessment of the K12 education morning inspection data is more accurate and efficient using the combined model.