UPB @ ACTI: Detecting Conspiracies using fine tuned Sentence Transformers

TL;DR

This paper presents a method using fine-tuned Sentence Transformers and data augmentation to effectively detect conspiracy theories online, achieving top performance in the ACTI @ EVALITA 2023 shared task.

Contribution

It introduces a novel approach combining pre-trained Sentence Transformers with data augmentation for conspiracy theory detection, outperforming existing systems.

Findings

Achieved 85.71% F1 in binary classification

Achieved 91.23% F1 in fine-grained classification

Secured first place in both sub-tasks

Abstract

Conspiracy theories have become a prominent and concerning aspect of online discourse, posing challenges to information integrity and societal trust. As such, we address conspiracy theory detection as proposed by the ACTI @ EVALITA 2023 shared task. The combination of pre-trained sentence Transformer models and data augmentation techniques enabled us to secure first place in the final leaderboard of both sub-tasks. Our methodology attained F1 scores of 85.71% in the binary classification and 91.23% for the fine-grained conspiracy topic classification, surpassing other competing systems.