Intelligent paper marking implementation method and system based on deep learning and computer program

Abstract

The invention belongs to the technical field of artificial intelligence, and discloses an intelligent paper marking implementation method and system based on deep learning, and a computer program. Thetest paper of the student is scanned, and test paper information is converted into picture information to be stored in a local computer hard disk; through optical character recognition, a natural scene text detection model is used for detecting the answer positions of subjective questions of examinees, and a text recognition model and a connection main sense time classification model are used forrecognizing text information of the answered contents of the examinees; through natural language processing, subjective question scoring is helped to be carried out by a volume batch person, similarity matching is carried out through word segmentation, spelling and grammar errors are found through semantic analysis of contexts, and the volume batch person is helped to give a score of the test paper. According to the method, the handwritten text can be highly accurately identified, and a volume holder can be helped to correctly score objective questions and subjective questions; the paper inspection time is greatly reduced, the objectivity of an examination is enhanced, and the labor of manpower resources is saved; and the cost loss caused by purchasing an expensive paper marking machine is reduced.

Description

technical field [0001] The invention belongs to the technical field of artificial intelligence, and in particular relates to a deep learning-based intelligent scoring implementation method, system, and computer program. Background technique [0002] At present, the existing technology in the industry is as follows: in the information age of rapid social development, the enrichment of teaching content on the one hand makes the learning tasks of students continue to increase, and on the other hand, the burden on teachers also increases. Especially in today's examination-oriented education situation has not completely changed, the high-efficiency method of rewriting papers has gradually become the voice of the majority of colleges and universities. In the current grading mode, the manual waiting for manual binding of examination papers, grading and circulation of examination papers has greatly delayed the grading process, and will produce many unnecessary human errors; even mac...

AI Technical Summary

Problems solved by technology

[0005] (1) In the current marking mode, manual binding of test papers, manual registration of scores and manual waiting for test paper circulation all delay the process of marking to a large extent, and will cause many unnecessary human errors

[0006] (2) Existing machine marking is not only expensive, but also can only play an auxiliary role in reviewing multiple-choice questions. Most of the subjective questions are still corrected by teachers, and the burden on teachers is still heavy

[0007] (3) The traditional manual marking method cannot timely shield the candidates' information, resulting in different evaluation standards for different teachers, and the difference in scores caused by subjective factors

Images