A Machine Learning SVM-Based Text Difficulty Assessment Model for Chinese Language International Education

As the rapid advancement of artificial intelligence technology and the growing demand for Chinese language learning, the assessment of text difficulty in Chinese language international education has become a core task for optimising the allocation of learning resources. To address the limitations of traditional assessment methods in terms of subjectivity, adaptability and large-scale automated processing. The study combines the two-way encoder language model with support vector machine to propose a novel text difficulty assessment model for Chinese international education. The findings denoted that the proposed text difficulty assessment model for Chinese international education has good precision and comprehensive quality. Its precision rate, recall rate and F1 value reach $95.8 \%, 96.3 \%$ and 95.8%, respectively. Besides, the comprehensive coverage of the model reaches 76.39%, indicating that the model is able to accurately predict the difficulty of about 76.39% of the textbooks, while the adjacency accuracy reaches 94.3%. The text difficulty assessment model proposed by the study is not only able to effectively capture the linguistic features and semantic information of the text, but also possesses strong generalisation ability and stability, and is able to provide an accurate and intelligent text difficulty assessment scheme for Chinese language international education.