Research on Graduation Thesis Evaluation Based on Hopfield Neural Network

Abstract

The evaluation of graduation thesis needs to consider from many aspects to make the final evaluation of each factor. The traditional evaluation method with weighted method has a loss of objectivity and lack of convincing. In view of the existing problems, it proposed that using Hopfield method to evaluate the graduation thesis. Through the existing evaluation results, the evaluation criteria were got. The evaluation criteria were stored in the network. When inputting the pre-evaluation thesis to the network, Hopfield neural network will run by itself. The pre-evaluation thesis will converge to the closest level. The experiments were carried out with 10 group's data. The experimental results show that this method can effectively evaluate the graduation thesis, and the evaluation results more rapid, rational and objective.