The usage of artificial Intelligence-empowered text analysis model with convolutional neural network in english reading.

With the increasing application of artificial intelligence (AI) in education, deep learning models have offered new solutions for personalized reading instruction. This study addresses the core issue of insufficient personalization in high school English reading instruction by systematically applying the Text Convolutional Neural Network (Text CNN) model in real classroom settings for the first time. A teaching-assistive framework is developed, integrating dynamic content recommendation with real-time feedback. Using multi-scale feature extraction and an attention mechanism based on Text CNN, the model automatically classifies reading materials, extracts key terms, and delivers personalized resources. A comparative experiment was conducted in a high school with 60 students, divided into an experimental group and a control group. The results showed that the experimental group, assisted by the Text CNN tool, improved their average reading comprehension scores from 66.2 to 79.6, outperforming the control group's 70.3. The model achieved classification accuracies of 93.5% and 94.2% for science and education texts, respectively, while reducing teachers' grading workload by 70%. However, this study was validated on a limited dataset of 2,000 texts, and the model's performance on unstructured texts (e.g., social media posts or legal documents) remains to be explored. Overall, this study provides empirical evidence for applying Text CNN in English instruction and lays a theoretical and practical foundation for developing and optimizing intelligent educational tools.