Sentiment Polarity as a Stylistic Feature for Attribution of English Prose

Authors

DOI:

https://doi.org/10.32473/flairs.38.1.138726

Keywords:

sentiment analysis, authorship attribution, BERT, VADER, TextBlob, prose, English, poetry, ensemble

Abstract

In this paper, we expand our exploration the usefulness of sentiment polarity as a stylistic feature for authorship attribution. We briefly describe our methodology and present the results from experiments using three English prose corpora. A comparison with attribution performed using traditional stylistic features on the same corpora is presented along with a brief comparison with prior results on sentiment-polarity-based attribution of poetry. We point out some limitations of our methodology and outline directions for further research.

Downloads

Published

14-05-2025

How to Cite

Ivanov, L. (2025). Sentiment Polarity as a Stylistic Feature for Attribution of English Prose. The International FLAIRS Conference Proceedings, 38(1). https://doi.org/10.32473/flairs.38.1.138726

Issue

Section

Special Track: Applied Natural Language Processing