Constructing New Media Student Management System in Colleges and Universities

Abstract

The emergence and widespread use of new media have led to its integration into people's daily life. Because they utilize new media so frequently, students need social support and attention when using it for student management. Concerns about college student growth, societal advancement, concord, and the management impact of colleges and universities are all relevant. In this manuscript, constructing new media student management system in colleges and universities (CNM-SMS-CU-SASGNN-PO) is proposed. Initially input data are gathered from Students' dropout and academic success dataset. To execute this, input data is pre-processed, Surface Normal Gabor Filter (SNGF) and data normalizing; then pre- processed image is fed to new media construction utilizing Structure-Aware Siamese Graph Neural Network (SASGNN) to construct new media student management system. In generally, SASGNNdoesn’t express adapting optimization strategies to determine optimal parameters to ensure construct new media student management system. Therefore, Parrot Optimizer (PO)is to optimize Structure-Aware Siamese Graph Neural Network which accurately constructs the student management system. Then the proposed CNM-SMS-CU-SASGNN-PO is implemented in Python and performance metrics analysis is done on metrics including Error Rate, NMSE (Normalised Mean Square Error), Accuracy, Root Mean Square Error (RMSE), and MAE (Mean Absolute Error). Performance of the CNM-SMS-CU-SASGNN-PO approach attains 18.27%, 23.65% and 32.60%  higher accuracy, 19.55%, 22.85% and 32.10%lower RMSE and 18.47%, 22.55% and 32.79% lower Error Rate when analysed through existing techniques like The design of a college student achievement management system based on the GA-BP network (CSA-MS-BPNN) and artificial neural network analysis of students' academic performance in virtual learning environments (APS-VLE-ANN) approaches, respectively, are the subjects of Application of New Media in Student Management from the Perspective of Deep Learning and Evaluation and Analysis of Practical Effects (NM-SMS-CNN)..