A Study On The Prediction Of Talent Demand And Training Matching Of SARIMA Model Based On Neural Network

With the growth of population,the positions of various companies tend to be saturated,and the requirements of enterprises for job seekers are increasingly high.The real-time post data of the recruitment platform reflects the needs of enterprises for job-seekers,which provides a good direction for the establishment of training policies in Colleges.In the outline of action for promoting the development of big data,big data has become a national strategy.Now,It is urgent to cultivate a group of composite talents with big data thinking and professional technology.In this paper,the research of "improved SARIMA model + Pearson correlation coefficient" is put forward to model,and forecast the demand of big data posts in Jiangsu Province from January 2016 to December 2019,and to match the demand forecast trend of enterprises with the professional supply of colleges.With the advantages of high fault tolerance,the model can be widely used in the practical application of recruitment demand and talent matching,so as to alleviate the phenomenon of disconnection between talent training and talent demand in the future.This paper subject has conducted research on the above issues,and the main work in the research process is as follows:In the aspect of enterprise recruitment demanding prediction model,obtained the demand for talent positions from a recruitment website based on Python.From the processed sequence characteristics,the SARIMA-BP model was selected to predict the changing trend of Jiangsu enterprises' demand for big data talents,and the prediction results are evaluated.This model has not yet been applied to the field of recruitment data prediction,but under the premise of a large base,the root mean square error is only 7.66.With high accuracy and feasibility,it provides data model support for subsequent matching research and problem discovery.In the study of matching degree research,sorted out the settings of big data majors in colleges in Jiangsu Province in the past eight years.The perspective of dataization objectively evaluates the current situation of talent supply and talent demand in colleges by Pearson correlation coefficient.In the survey of the development status of the big data education industry,the Lorentz curve and Gini coefficient are used to match the current status of new big data majors in each province with the amount of colleges,which provides data support for the professional settings.In view of the matching research results and the current situation of the development of big data education industry,in order to solve the problems existing in the construction of big data majors andthe needs of posts in Colleges in Jiangsu Province,this paper puts forward suggestions under the pattern of "one body,two wings and one tail".