Computational Authorship Attribution in the Children’s Tales of Oscar and Constance Wilde: The Case of "The Selfish Giant"

2026-06-25 · Source: Paper Index on ACL Anthology · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics · Depth: Expert, quick

Summary

A computational authorship attribution study investigated children's stories by Oscar and Constance Wilde, specifically examining "The Selfish Giant." Researchers utilized both supervised methods, including SVM with string kernel, and unsupervised techniques like Hierarchical Clustering via Rank Distance. Their analysis revealed a strong stylistic similarity between "The Selfish Giant" and Constance Wilde's stylometric profile, suggesting the story could be a collaborative effort. Furthermore, the study explored the efficacy of Large Language Models (LLMs) in authorship attribution through Perplexity. The results confirmed the distinct stylistic fingerprints of both authors within the corpus, demonstrating their individual styles are separable despite shared genre and publication period.

Key takeaway

For research scientists or digital humanists investigating literary authorship, this study highlights the power of combining traditional stylometric methods with Large Language Model perplexity. You should consider integrating LLM-based analysis into your attribution workflows, especially when dealing with stylistically similar authors or collaborative works. This multi-faceted approach offers a more robust framework for identifying distinct authorial fingerprints and resolving complex attribution cases.

Key insights

Stylometric analysis, including LLM perplexity, suggests "The Selfish Giant" may be a collaboration between Oscar and Constance Wilde.

Principles

• Stylometric profiles can differentiate authors even in shared genres.
• LLMs can complement traditional stylometric methods.
• Combined methods enhance authorship attribution robustness.

Method

The study combines supervised (SVM with string kernel) and unsupervised (Hierarchical Clustering via Rank Distance) stylometric methods, augmented by LLM perplexity analysis for authorship attribution.

In practice

• Apply stylometric analysis to resolve disputed literary authorship.
• Use LLM perplexity as an authorship attribution tool.
• Integrate diverse computational methods for robust analysis.

Topics

• Computational Stylistics
• Authorship Attribution
• Large Language Models
• Perplexity
• Digital Humanities
• SVM
• Hierarchical Clustering

Best for: NLP Engineer, AI Scientist, Research Scientist

Related on AIssential

• 'The Order in the Horse's Heart': A Case Study in LLM-Assisted Stylometry for the Discovery of Biblical Allusion in Modern Literary Fiction — A dual-track LLM pipeline effectively detects biblical allusions in literature, cross-validated against scholarship.
• Authorship Attribution in Multilingual Machine-Generated Texts — Multilingual authorship attribution for LLM-generated text is complex, with current monolingual methods showing limited cross-lingual transferability across diverse language families.
• ‘But this one was so . . . male.’ A Corpus-Based and LLM-Augmented Analysis of Language and Gender Bias in Barbie — Combining corpus linguistics with LLMs effectively reveals subtle gender biases in film dialogue.
• Three Buddhist Vocabularies: Computational Stylometry of the English Pali Canon across Sutta, Vinaya, and Abhidhamma — Computational stylometry reveals distinct lexical patterns and significant translation shifts across Buddhist canonical texts.
• How Far Can Classical NLP Go? From Bag-of-Words to Stacking on Spooky Author Identification — Classical NLP, with careful feature engineering and ensemble stacking, excels at stylistic text classification.
• Context Graphs: Free Association, Consistency, Narratives & Truth — LLMs function via free association on narrative graphs, trading consistency for broader, stochastically generated outputs.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Paper Index on ACL Anthology.