Method for evaluating poor students in colleges and universities

Abstract

The present invention provides a method for assessing impoverished students in colleges and universities, which includes the following steps: ① Obtaining parameters: obtaining the measured values ​​of ten indicators in the data of college students who have been evaluated and recorded in the system, and the ten indicators are respectively the monthly average family income FI, Per capita monthly income and expenditure ratio FIOR, the ratio of labor force members to family members FMR, special family type FT, family member health status FH, source of tuition ST, source of living expenses SC, average daily food consumption SE, class democratic assessment SS and failure status SCS; ②model discrimination; ③obtain results. The present invention adopts the method of the forward supervised neural network algorithm model, and selects the indicators more reasonably, so that the accuracy can exceed 93.3%, avoiding the situation of over-fitting to a large extent, and the accuracy rate is ideal, so that the school It can assess poor students more truly, fairly, conveniently and efficiently.

Description

technical field [0001] The invention relates to a method for evaluating impoverished students in colleges and universities. Background technique [0002] In order to reduce the financial burden of poor families, minimize the study of students who are troubled by financial problems. The party and the state have always attached great importance to the work of supporting poor students and implemented such as "Notice on Funding Students with Special Difficulties in Colleges and Universities", "Notice on Further Strengthening the Work of Colleges and Universities to Support Students with Economic Difficulties" and "The Ministry of Education and the Ministry of Finance on Conscientious Some policies, such as the Guiding Opinions on Doing a Good Job in the Identification of Students from Economically Difficult Families in Colleges and Universities, have benefited poor students, reducing the pressure of their financial burden and allowing them to study with peace of mind. According...

AI Technical Summary

Problems solved by technology

[0004] The indicators used in the above studies are higher than the actual actual operation or the acquisition of indicators is less cost-effective, so that the method used cannot be implemented well, and there is an overfitting problem in the use of decision trees, and there is no comparison in the consistency judgment of the matrix in the gray comprehensive evaluation. scientific guidance

However, there are few studies on actual and easy-to-obtain evaluation indicators for poor students

Images