Matching Prediction of Teacher Demand and Training Based on SARIMA Model Based on Neural Network

Abstract

This study introduces the ‘SARIMA Improved Model + Pearson Correlation Coefficient' approach to predict the demand for big data jobs in Jiangsu Province schools from January 2016 to December 2019. It also explores the matching between demand and supply in universities. The model is fault-tolerant, offers fast predictions, and addresses the disconnect between college talent training and teacher demand. The SARIMA-BP model predicts the trend of big data teacher demand in Jiangsu Province. The model, though untested in recruitment data prediction, with a large database, achieves root mean square error of 7.66, indicating high precision and reliability. Based on matching research and the local big data education industry in Jiangsu Province, countermeasures and suggestions are presented under the “one body, two wings, and one tail” framework. This concise summary highlights the research's core components and objectives.