Evaluation of University Project Based on Partial Least Squares and Dynamic Back Propagation Neural Network Group

In this paper, aiming at the more objective evaluation of university scientific research projects, an evaluation model based on partial least squares and dynamic back propagation neural network group (PLS-DBPG) algorithm is put forward. First, the reasonable index system of project evaluation is designed. For evaluation indexes, the PLS algorithm is used to reduce the feature dimension of original data. As the input of BP neural network, low dimensional data can simplify network structure. Then in view of the slow convergence of traditional BP algorithm, a dynamic adaptive BP neural network group algorithm is proposed. By introducing dynamic proportion factors, the update of each BP neural network is affected by both itself and the optimal network. And, the contribution of each network for group study is constantly adjusted. Finally, in this paper, the simulation results show that the improved algorithm for evaluating the project proposed has valuable applications with the merits of the fast convergence and high precision.

Abstract licence: CC BY-NC-ND

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.