Adaptive Graph Neural Network with Cross-Source Attention and Dynamic Hyperparameter Regulation for Structured Modeling of Multi-Source Literary Corpora
Abstract
This paper proposes an adaptive graph neural network framework integrating cross-source attention and dynamic hyperparameter regulation for structured modeling of multi-source literary corpora. The dataset includes four genres—ancient books, modern novels, online literature, and bilingual translations—comprising about two million tokens across 18,000 chapters. Experiments on an NVIDIA RTX 3090 show that the proposed model achieves an average accuracy of 91.7%, macro F1 of 90.2%, and RMSE of 0.142, outperforming the fixed-parameter baseline by approximately 4%. The convergence speed improves by 20%, and robustness is maintained under small-scale, noisy, and cross-language conditions. Ablation results confirm the independent contribution of each module. The proposed mechanism achieves an effective balance between performance and efficiency, offering a reproducible and scalable approach for digital humanities and cross-disciplinary text analysis.References
Wang C, Liu X, Chen Z, et al. DeepStruct: Pretraining of language models for structure prediction[J]. arXiv preprint arXiv:2205.10475, 2022.https://doi.org/10.48550/arXiv.2205.10475
Deng S, Mao S, Zhang N, et al. SPEECH: Structured prediction with energy-based event-centric hyperspheres[J]. arXiv preprint arXiv:2305.13617, 2023.https://doi.org/10.48550/arXiv.2305.13617
Li X, Li J. Angle-optimized text embeddings[J]. arXiv preprint arXiv:2309.12871, 2023.https://doi.org/10.48550/arXiv.2309.12871
Nguyen T, Zhang Q, Yang B, et al. Predicting from Strings: Language Model Embeddings for Bayesian Optimization[J]. arXiv preprint arXiv:2410.10190, 2024.https://doi.org/10.48550/arXiv.2410.10190
Perri V, Qarkaxhija L, Zehe A, et al. One graph to rule them all: Using nlp and graph neural networks to analyse tolkien's legendarium[J]. arXiv preprint arXiv:2210.07871, 2022.https://doi.org/10.48550/arXiv.2210.07871
Zhao H, Xie J, Yan Y, et al. A corpus for named entity recognition in Chinese novels with multi-genres[J]. arXiv preprint arXiv:2311.15509, 2023.https://doi.org/10.48550/arXiv.2311.15509
Ke S, Montiel Olea J L, Nesbit J. Robust machine learning algorithms for text analysis[J]. Quantitative Economics, 2024, 15(4): 939-970.https://doi.org/10.3982/QE1825
Huang S, Yang K, Qi S, et al. When large language model meets optimization[J]. Swarm and Evolutionary Computation, 2024, 90: 101663.https://doi.org/10.1016/j.swevo.2024.101663
Das L, Ahuja L, Pandey A. A novel deep learning model-based optimization algorithm for text message spam detection: L Das et al[J]. The Journal of Supercomputing, 2024, 80(12): 17823-17848.https://doi.org/10.1007/s11227-024-06148-z
Wang L, Yang N, Huang X, et al. Improving text embeddings with large language models[J]. arXiv preprint arXiv:2401.00368, 2023.https://doi.org/10.48550/arXiv.2401.00368
Wang K, Ding Y, Han S C. Graph neural networks for text classification: A survey[J]. Artificial intelligence review, 2024, 57(8): 190.https://doi.org/10.1007/s10462-024-10808-0
Xu Y, Mao C, Wang Z, et al. Semantic-enhanced graph neural network for named entity recognition in ancient Chinese books[J]. Scientific Reports, 2024, 14(1): 17488.https://doi.org/10.1038/s41598-024-68561-x
Lu Z, Xie Q, Wang B, et al. Word grounded graph convolutional network[J]. arXiv preprint arXiv:2305.06434, 2023.https://doi.org/10.48550/arXiv.2305.06434
Dai Q. Construction of English and American literature corpus based on machine learning algorithm[J]. Computational Intelligence and Neuroscience, 2022, 2022(1): 9773452.https://doi.org/10.1155/2022/9773452
Sobchuk O, Šeļa A. Computational thematics: comparing algorithms for clustering the genres of literary fiction[J]. Humanities and Social Sciences Communications, 2024, 11(1): 1-12.https://doi.org/10.1057/s41599-024-02933-6
Tripto N I, Ali M E. The word2vec graph model for author attribution and genre detection in literary analysis[J]. arXiv preprint arXiv:2310.16972, 2023.https://doi.org/10.48550/arXiv.2310.16972
Hatzel H O, Stiemer H, Biemann C, et al. Machine learning in computational literary studies[J]. it-Information Technology, 2023, 65(4-5): 200-217.https://doi.org/10.1515/itit-2023-0041
Yang C, Wang X, Lu Y, et al. Large language models as optimizers[C]//The Twelfth International Conference on Learning Representations. 2023.https://doi.org/10.48550/arXiv.2309.03409
DOI:
https://doi.org/10.31449/inf.v49i27.12021Downloads
Published
How to Cite
Issue
Section
License
I assign to Informatica, An International Journal of Computing and Informatics ("Journal") the copyright in the manuscript identified above and any additional material (figures, tables, illustrations, software or other information intended for publication) submitted as part of or as a supplement to the manuscript ("Paper") in all forms and media throughout the world, in all languages, for the full term of copyright, effective when and if the article is accepted for publication. This transfer includes the right to reproduce and/or to distribute the Paper to other journals or digital libraries in electronic and online forms and systems.
I understand that I retain the rights to use the pre-prints, off-prints, accepted manuscript and published journal Paper for personal use, scholarly purposes and internal institutional use.
In certain cases, I can ask for retaining the publishing rights of the Paper. The Journal can permit or deny the request for publishing rights, to which I fully agree.
I declare that the submitted Paper is original, has been written by the stated authors and has not been published elsewhere nor is currently being considered for publication by any other journal and will not be submitted for such review while under review by this Journal. The Paper contains no material that violates proprietary rights of any other person or entity. I have obtained written permission from copyright owners for any excerpts from copyrighted works that are included and have credited the sources in my article. I have informed the co-author(s) of the terms of this publishing agreement.
Copyright © Slovenian Society Informatika
