Model and research on score evaluation model of graduation thesis

The score evaluation of graduation thesis is an important means to inspect, evaluate and promote graduation thesis. In this paper, by using the significance test method of the multivariate statistical analysis, we examined the impact significance of each evaluation index on graduation thesis total evaluation result, based on this to select the appropriate evaluation index and to construct a score evaluation model. By means of BP neural network, we constructed the non-linear model between the graduation thesis evaluation indices and overall mark, and finally formed an effective score evaluation method on graduation thesis.