A deep learning-based college student credit assessment method and device

Abstract

The invention discloses a college student credit assessment method and device based on deep learning. The method comprises the steps of A, obtaining credit data of a to-be-assessed student; B, converting text information in the credit data sample into word vectors by using a word2vec framework, and processing the word vectors corresponding to the to-be-evaluated students by using a pre-trained LSTM model to obtain text characteristic index data; C, performing feature assignment on other credit data except the text information in the credit data of the to-be-evaluated student, and performing normalization processing to obtain non-text feature index data; And D, using a pre-trained BP neural network to map the student text feature index data and the non-text feature index data as continuousinput and continuously output to obtain a target credit assessment result of the to-be-assessed student. By applying the embodiment of the invention, the efficiency and accuracy of credit assessment for students can be improved.

Description

technical field [0001] The present invention relates to a method and device for credit evaluation of college students based on deep learning, and more particularly to a method and device for credit evaluation of college students based on deep learning. Background technique [0002] In recent years, there have been frequent incidents of untrustworthy college students in their studies, academia, and employment, such as academic fraud, graduate student loan arrears, etc. The lack of credit of college students not only damages the overall image of college students, is not conducive to the healthy growth of college students themselves, but also It also has a bad influence on the society, causing the public to have many doubts about university education. By analyzing the information about the study and life of college students, it is helpful to form a relatively complete credit evaluation method that is in line with the characteristics of college students. Therefore, the research ...

AI Technical Summary

Problems solved by technology

The method based on manual judgment relies on the mutual evaluation among classmates and the comprehensive evaluation results of counselors and teachers. This method is based on the subjective feelings of direct interaction between individuals, making it difficult for the evaluation results to reflect the true credit level. There is a discrepancy between the obtained assessment results and the real results

The multi-attribute weight analysis method is based on the student achievement information database, and the weighted average of student achievement and credits is used as the result of student credit evaluation. This method only uses student achievement as the basis for credit evaluation, which is too one-sided and does not make enough use of other behavioral information of students. Poor rating

[0004] In summary, there is a technical problem in the existing technology that the credit evaluation results for students are not accurate enough

Images