A big data-based research method for digitizing ancient literary texts

Abstract

This paper focuses on the innovative application of big data technology in the field of digitization of ancient literary texts, and puts forward a new mode and method of using big data technology to realize the whole-process design and implementation, in-depth analysis and excavation of ancient literary texts, in view of the inadequacy of the current digitization of ancient literary texts in terms of comprehensiveness and depth. The research practice shows that with the help of big data technology, not only has remarkable results been achieved in data collection and integration, text processing standardization, and efficient data management, but also in the application of specific technological tools such as Bayesian Intelligent Error Correction Algorithm, bilstm-CRF Ancient Literature Text Structuring and Semantic Annotation, and BERT Sentiment Connotation Analysis Model, which demonstrates excellent performance. The results show that by adopting advanced digitization research methods, the recognition accuracy of ancient texts, literary references, author information and publication year reaches 98%, 97%, 97%, and 98.5%, respectively, and after standardization and structuring, the heterogeneity of each data item is greatly reduced, the degree of structuring is significantly improved, and the data usability and accessibility are also enhanced. Meanwhile, the evaluation data of error correction algorithm efficacy, structuring function, and sentiment analysis function show that the model is outstanding in reducing the error rate, improving the degree of structuring, and performing sentiment analysis, especially for the processing of ancient texts and literary references, the error correction effect and the accuracy rate of sentiment analysis are especially obvious. In summary, this study has successfully constructed a comprehensive and in-depth digitization system of ancient literary texts through big data technology, which not only greatly improves the quality and usability of the data, but also broadens the perspectives and dimensions of ancient literature research.