As a treasure of Chinese culture, Tang Dynasty literature carries rich cultural values, but lacks a systematic evaluation model. Traditional evaluation methods are highly subjective and difficult to comprehensively reflect the multidimensional value of Tang Dynasty literature. This study constructs a comprehensive evaluation model of the cultural value of Tang Dynasty literary works based on the improved Transformer model, and realizes the automatic evaluation of the cultural value of Tang Dynasty literary works through the technical improvements of the feature participle layer, the output participle CLS Token and the Pre-Norm structure. The study establishes an evaluation system containing 21 evaluation indexes from six aspects: historical dimension, ideological dimension, artistic dimension, humanistic dimension, cultural inheritance dimension and social function dimension. The experiment uses 1200 sample data, of which 1000 are used for training and 200 for testing. The results show that the improved Transformer model achieves an accuracy of 85% on the test set, with a precision of 1.0 and an F1 score of 0.862, which is better than the traditional CNN model and LSTM model. In further validation experiments, the model achieves an overall prediction accuracy of 92.65%, which is significantly higher than the CNN model’s 80.65% and the LSTM model’s 80.01%. Factor analysis showed that the cumulative variance contribution rate of the six main factors reached 83.79%, which can comprehensively reflect the cultural value of Tang Dynasty literary works. This study provides an objective and efficient quantitative method for evaluating the cultural value of Tang Dynasty literary works.
