This paper combines N-gram models, LDA topic models, TF-IDF algorithms, and clustering analysis methods, and based on the LIA-BiLSTM text sentiment analysis model, collects, organizes, and analyzes a large amount of Chinese language and literature texts. This paper uses big data technology to deeply explore the educational value of traditional culture in Chinese language and literature and discusses its application path in the field of vocational education. Research indicates that the improved LIA-BiLSTM model developed in this study outperforms other models, with an AUC value of 0.9779, approaching 1. Traditional cultural elements in Chinese language and literature help enhance students’ sense of cultural confidence, belonging, achievement, and responsibility. By constructing an integrated educational model of “course integration—cultural immersion—practical experience—evaluation feedback,” this approach not only effectively preserves China’s excellent traditional culture but also infuses vocational education with new cultural significance.