Research on Automatic Precise Scoring Model and Method for Mid-term English Examination

Automatic scoring of English essays has always been a hot topic in language teaching and AI research, especially in recent years with the emergence of large language models, which have provided important tool support for English essay automatic scoring research. However, as an important selection exam, the research on automatic scoring of English essays in domestic middle school entrance exams is still basically at the exploratory stage, and there is a clear gap with the automatic scoring practices of important international exams. In order to further optimize the model and application research of automatic scoring of English essays in important selection exams, this paper, based on reference to relevant research at home and abroad, and combined with some exploratory practices of automatic scoring of English essays in the middle school entrance examination in a certain city, proposed an heterogeneous scoring model and method, and through the comparison of the model with the manual scoring data in the training set and the test set, the precision of the model in the same theme test set can reach the precision of manual scoring. The relevant framework, model and algorithm can lay a foundation for the subsequent large-scale practical application.