The evaluation of university management performance using the CS-RBM algorithm

Abstract

Amidst the ongoing higher education reforms in China, the escalated investments in colleges and universities underscore the need for an effective assessment of their performance to ensure sustainable development. However, traditional evaluation methods have proven time-consuming and labor-intensive. In response, a novel approach called CS-RBM (Crow Search Restricted Boltzmann Machine) prediction algorithm has been proposed for the educational management of these institutions. By integrating the CS algorithm and an enhanced RBM algorithm, this method facilitates the scoring of project performance indicators, bolstered by insights from user evaluation form reports. The comprehensive project performance is ultimately derived from this combination. Comparative analysis with the standard particle swarm optimization algorithm on public data sets demonstrates a remarkable 45.6% reduction in prediction errors and an impressive 34.7% increase in iteration speed using the CS-RBM algorithm. The accuracy of the tested data set surpasses 98%, validating the efficacy of the CS-RBM algorithm in achieving precise predictions and effective assessments. Consequently, this innovative approach exhibits promising potential for expediting and enhancing the performance evaluation of colleges and universities, contributing significantly to their sustainable development.