Prediction Research On Student Status Change For Students Based On Improved Neural Network Algorithm

At present,the management of student status in universities has been basically achieved systematization and informatization,but the management is still involved in all aspects,especially the management of the change of student status.The research shows that there is a large number of students having the change of status in universities every year,but most of these cases are recorded and analyzed through table or system,the operation is tedious,and it is difficult to find some of the potential information and patterns,some students are even unable to make sure whether they can finish their studies properly.Therefore,it is an important content for the management of universities to analyze and predict the change of student status effectively and timely,assess the accuracy of the prediction,and realize the early warning of students status change with methods of technology.In view of this situation,this paper attempts to propose a prediction model for the change of student status based on back propagation neural network algorithm.In order to achieve the goal,and ultimately provide assistant reference for university educational administration staff,this paper first clears the data features of the data set,extracts the eigenvalues,and to form an experimental training data set after data preprocessing.Secondly,by designing the topological structure of the BP neural network,defining the relevant parameters,selecting the appropriate activation function and the network training mode,the prediction model for the change of student status based on back propagation neural network is constructed.Then on the basis of the classical network algorithm,this paper uses the principal component analysis and sensitivity analysis to optimize the prediction model,in order to improve the efficiency of the model training and accuracy of the model prediction.Finally,the above network prediction models are evaluated and verified by actual data.On the one hand,it verifies the accuracy of the model in the prediction of student status change,on the other hand,it illustrates the practical application effect of the model,and proves the effectiveness of the model in the management of student status.In this paper,a prediction model for the change of student status based on neural network algorithm is constructed,and the model is optimized.It is verified that the optimized model can effectively predict the change of student status,and the prediction accuracy can reach nearly 89%.Compared with the non-optimized model,the prediction accuracy is increased by 5%.In the end,the optimized model is applied to the prediction of the 2016 grade student status change in universities,and the result shows that the prediction model can effectively assist the management of student status in making decisions.