Study of Argumentative Writing and Narrative Methods for English Majors based on Graph Neural Networks

The main purpose of this study is to investigate the differences between narrative writing errors and argumentative writing errors of Chinese learners by using the argumentative writing and narrative writing methods of English majors based on graph neural networks. Using graph neural networks and the error analysis method described by Corder (1981), our research classification method is consistent with the theoretical framework of Burt and Kiparski. Our analysis aims to identify the different nature and characteristics of the errors present in these two components. The correlation between the prevalence of global and local errors, as well as their relationship to different types of writing, is carefully examined through graph neural networks. The findings ultimately suggest a targeted teaching approach tailored specifically to address common errors in different writing styles. This method is expected to significantly improve the writing level of English majors.