EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

An AI-Driven Computational Framework for Intelligent Literary Analysis, Natural Language Processing, Semantic Interpretation, and Data-Driven Decision Making in Digital Humanities

Dr A. Devasena, Robert Sunil F, Dr.P.Suria Thilagam, Sagar M Baligidad, Dr. Ravi Kumar Meena, Mousumi Paul

Computer and Decision Making: An International Journal · 08.11.2026 · 3 (8S) · 245-256

DOI: 10.1922/comdem.v3i8s.19174 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer and Decision Making: An International Journal, online erschienen am 08.11.2026. Verfasst von Dr A. Devasena und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Digital humanities have started to adopt artificial intelligence, but literary analysis, semantic interpretation, classification and scholarly decisions support are not necessarily considered together. This work proposes an integrated multi-view computational framework that integrates stylometric indicators, TF-IDF lexical features, Sentence-BERT embeddings, BERTopic thematization, supervised machine learning, ablation analysis, UMAP visualization, and SHAP-based method of explainability. The analysis was performed on a set of 5,711 fiction books in the English language, all of which are public domain, from the Artificial Relationships in Fiction / Project Gutenberg Fiction Corpus, in six literary genres. The proposed hybrid model obtained accuracy of 0.871, macro-precision of 0.868, macro-recall of 0.859, macro-F1 of 0.863 and macro-ROC-AUC of 0.967 which outperformed all the evaluated baselines. Ablation results showed that semantic embedding yielded the best individual representation while the full feature combination yielded the best results. The framework also translated the concept of “semantic coherence” and “stylistic distinctiveness” and “thematic structure” and “confidence” and “outlier behaviour” into understandable and interpretable decision support evidence for digital-humanities research.

Zitieren

APA

Devasena, D. A., F, R. S., Thilagam, D., Baligidad, S. M., Meena, D. R. K., & Paul, M. (2026). An AI-Driven Computational Framework for Intelligent Literary Analysis, Natural Language Processing, Semantic Interpretation, and Data-Driven Decision Making in Digital Humanities. Computer and Decision Making: An International Journal, 3(8S), 245-256. https://doi.org/10.1922/comdem.v3i8s.19174

Vancouver

Devasena DA, F RS, Thilagam D, Baligidad SM, Meena DRK, Paul M. An AI-Driven Computational Framework for Intelligent Literary Analysis, Natural Language Processing, Semantic Interpretation, and Data-Driven Decision Making in Digital Humanities. Computer and Decision Making: An International Journal. 2026;3(8S):245-256. doi:10.1922/comdem.v3i8s.19174

BibTeX
@article{Devasena2026,
  author = {Devasena, Dr A. and F, Robert Sunil and Thilagam, Dr.P.Suria and Baligidad, Sagar M and Meena, Dr. Ravi Kumar and Paul, Mousumi},
  title = {An AI-Driven Computational Framework for Intelligent Literary Analysis, Natural Language Processing, Semantic Interpretation, and Data-Driven Decision Making in Digital Humanities},
  journal = {Computer and Decision Making: An International Journal},
  year = {2026},
  volume = {3},
  number = {8S},
  pages = {245--256},
  doi = {10.1922/comdem.v3i8s.19174}
}
Kostenlos starten →