UPB @ ACTI: Detecting Conspiracies using fine tuned Sentence Transformers
This work addresses the problem of identifying conspiracy theories for improving information integrity and societal trust, but it is incremental as it applies existing methods to a new dataset.
The paper tackled conspiracy theory detection in online discourse, achieving first place in the ACTI @ EVALITA 2023 shared task with F1 scores of 85.71% for binary classification and 91.23% for fine-grained topic classification.
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.