Research On The Comprehensive Quality Evaluation Of College Students Based On Artificial Intelligence

Evaluation for college students'comprehensive quality is extremely important, as it relates to the students'development, school education and teaching reform and the ability of employing unit to obtain high-quality talents. To this end, this paper researched and established a set of comprehensive quality evaluation system for college students of institutions combined with arts and engineering from multi-subject perspective, integrating qualitative and quantitative analysis and satisfying the demand, and made a prediction for the students'occupations.This paper studied the factors affecting the comprehensive quality of college students of institutions combined with arts and engineering, established comprehensive quality evaluation indicator system for college students and conducted a quantitative analysis for each indicator, and determined the weight of each indicator leveraging the Analytic Hierarchy Process (AHP). In order to solve the problems of ambiguity and uncertainty of students'quality, fuzzy comprehensive evaluation is introduced in this paper to calculate the scores of comprehensive quality and sub-item quality modules for each student, however, the membership function and fuzzy rules cannot be automatically generated and adjusted when the fuzzy mathematics is applied to evaluate the students'comprehensive quality due to the complicated computing process. Therefore, this paper designed a new mathematical model for comprehensive evaluation, brought forward a combination algorithm named fuzzy neural network algorithm and established comprehensive quality fuzzy neural network model for college students using the respective advantages of fuzzy analysis and neural network in the evaluation. Moreover, in order to speed up the network convergence, the clustering analysis is applied in the training to cluster the value of each indicator. With the higher precise calculation, this method can realize the intellectualization for comprehensive quality; avoid the shortcomings in the fuzzy evaluation to better evaluate the students'quality. However, the speed of network convergence has not been fundamentally improved, in order to solve this problem, similarity measurement method in fuzzy mathematics is applied to improve the hidden layer nodes of the model, and the experiment has proved that the model has sped up the network convergence and optimized the network structure.This paper analyzed the advantages of the combination of wavelet analysis and neural network for the occupation prediction, established occupation prediction model based on the wavelet neural network, classifying the occupation into four categories and conducting training for network respectively. This model featured good simulation, higher precise prediction and short training period, and the prediction results provided the powerful basis for the students to comprehensively understand their own occupations.