Reddit is all you need: Authorship profiling for Romanian

TL;DR

This paper introduces a novel Romanian Reddit-based corpus for authorship profiling, demonstrating how LLMs can infer demographic and personal traits from social media texts, and providing a foundation for future NLP research in this area.

Contribution

It creates the first Romanian social media corpus annotated with author traits and evaluates LLMs for authorship profiling, advancing NLP capabilities in this language.

Findings

Successfully built a 23k+ sample Romanian Reddit corpus

Demonstrated LLMs can infer demographic traits from social media texts

Released resources publicly for further research

Abstract

Authorship profiling is the process of identifying an author's characteristics based on their writings. This centuries old problem has become more intriguing especially with recent developments in Natural Language Processing (NLP). In this paper, we introduce a corpus of short texts in the Romanian language, annotated with certain author characteristic keywords; to our knowledge, the first of its kind. In order to do this, we exploit a social media platform called Reddit. We leverage its thematic community-based structure (subreddits structure), which offers information about the author's background. We infer an user's demographic and some broad personal traits, such as age category, employment status, interests, and social orientation based on the subreddit and other cues. We thus obtain a 23k+ samples corpus, extracted from 100+ Romanian subreddits. We analyse our dataset, and finally, we fine-tune and evaluate Large Language Models (LLMs) to prove baselines capabilities for authorship profiling using the corpus, indicating the need for further research in the field. We publicly release all our resources.