Identifying Scientists on X

2026-09-25

AI Summary

This study addresses the urgent need to automatically identify scientist accounts on the X platform amid the proliferation of network science discourse and the disruption of knowledge order. We propose an ensemble classification framework that integrates a Random Forest model leveraging linguistic features with a contrastive learning fine-tuned DeBERTa model. By jointly modeling user profiles and tweet texts, this approach effectively distinguishes scientists from non-scientists. Experimental results demonstrate that the proposed ensemble achieves an F1 score of up to 0.96 across two benchmark datasets, significantly outperforming individual baseline models (F1=0.88). Furthermore, we construct and release a dedicated annotated dataset, providing essential data resources and methodological support for expert identity recognition research in computational social science.

Abstract

With the growing importance of science-related discourse on the Web and the erosion of the classical knowledge order, it is important to identify different user groups, such as scientists, automatically. This work proposes an approach for identifying scientists and non- scientists on X/Twitter based on their user biographies and tweets. We show that we are able to classify accounts as scientists and non- scientists on two different datasets, reaching an F1 score of up to 0.88 using Random Forests with linguistic features and up to 0.96 using a contrastively fine-tuned DeBERTa model in an ensemble setup. Furthermore, we provide two datasets with X users labeled as scientists or non scientists and their respective tweets and user biographies.

Problem

• Scientist Identification
• User Classification
• Social Media
• X/Twitter

Innovation

• Scientist Identification
• Contrastive Fine-tuning
• DeBERTa
• Ensemble Learning
• Social Media Mining