Automatic Test Scoring System Based on Deep Learning Technology

The traditional examination papers generally use manual marking, which is a very time-consuming and inefficient marking method. If teachers manually correct students' examination papers every time, not only the workload is heavy, but also due to the influence of subjective factors, they will not be able to objectively and impartially reflect the students' learning effect. If automatic scoring can be realized, it will greatly reduce the workload of teachers, improve the accuracy and objectiveness of scoring, and thus improve the quality of teaching. Therefore, solving the automatic scoring of examination papers is a key problem that needs to be solved urgently at present. In this paper, the BP neural network model in the deep learning technology is used to establish an automatic examination scoring system, which optimizes the information management performance of the system.