Development and Effect Evaluation of Ancient Chinese Reading Comprehension System Assisted by Artificial Intelligence

Reading comprehension of ancient Chinese faces many challenges, including the inefficiency of traditional learning methods. Therefore, this study developed an artificial intelligence-based ancient Chinese reading comprehension assistant system that combines natural language processing technology and deep learning models to improve the automation level of ancient Chinese processing. The system adopts a hierarchical architecture: the data preprocessing layer cleans the text and realizes word segmentation, and word segmentation accuracy is improved using a bidirectional long short-term memory + conditional random field model combined with knowledge enhancement technology. With the help of the Stanza library, dependency syntax analysis is carried out, and semantic understanding is enhanced through bidirectional encoder representations from transformers fine-tuning and the incorporation of a knowledge map. Experiments show that the system has obvious advantages in the accuracy and practicality of ancient Chinese prose processing. Future work may further improve the automatic analysis of cultural metaphors through small sample learning and multi-modal fusion.