Method for training test question scoring model and test question scoring method and device

Abstract

The invention discloses a method for training a test question scoring model and a test question scoring method and device. The method comprises the steps of obtaining student answer sheets corresponding to test questions in a question bank, wherein the student answer sheets comprise answer results corresponding to the test questions in the question bank; according to the answering result, the score of the answer sheet of the student is determined through a second test question scoring model, and the second test question scoring model is obtained by training a second neural network model embedded into the first test question scoring model through scoring data of sample test questions in a question bank under the condition that parameters of the first test question scoring model are kept unchanged. The first test question scoring model is obtained by pre-training the first neural network model by using the scoring data of the sample test questions outside the question bank. According to the method, the effectiveness of scoring the specific test questions by the scoring model can be ensured, and the training cost, the maintenance cost and the deployment cost are reduced.

Description

technical field [0001] The present application relates to the field of computer technology, and in particular to a method for training a test question scoring model, a test question scoring method, and an apparatus. Background technique [0002] In recent years, with the rapid development of computer technology and artificial intelligence, all walks of life have the need to replace labor with machines, and there is also a strong demand in the domestic education field. [0003] The use of machines for automatic question grading can greatly reduce the burden on teachers, so it is more and more widely used. For example, composition is a necessary question type for students’ Chinese and English tests. Currently, there are some automatic composition scoring methods in test scoring scenarios. The general scoring model of deep learning is used in the scene, but no matter which scoring model is used, the effect and efficiency of training the automatic scoring model need to be impro...

AI Technical Summary

Problems solved by technology

For example, composition is a necessary question type for students in Chinese and English exams. At present, there are some automatic scoring methods for composition in the test scoring scene, for example, using the deep learning calibration scoring model in the large-scale test scoring scene, and in the classroom test scoring The general scoring model of deep learning is used in the scene, but no matter what kind of scoring model it is, the effect and efficiency of training the automatic scoring model need to be improved

Images