Overview of PAN 2023: Authorship Verification, Multi-author Writing Style Analysis, Profiling Cryptocurrency Influencers, and Trigger Detection - Extended Abstract

Janek Bevendorff, Mara Chinea-Rios, Marc Franco-Salvador, Annina Heini, Erik Körner, Krzysztof Kredens, Maximilian Mayerl, Piotr Pezik, Martin Potthast, Francisco Rangel, Paolo Rosso, Efstathios Stamatatos, Benno Stein, Matti Wiegmann, Magdalena Wolska, Eva Zangerle

Abstract: The paper gives a brief overview of the four shared tasks organized at the PAN 2023 lab on digital text forensics and stylometry to be hosted at the CLEF 2023 conference. The general goal of the PAN lab is to advance the state-of-the-art in text forensics and stylometry while ensuring objective evaluation of new and established methods on newly developed benchmark datasets. PAN’s tasks cover four areas of digital text forensics: author identification, multi-author analysis, author profiling, and content analysis. Some tasks follow up on past editions (cross-domain authorship verification, multi-author writing style analysis) and some explore novel ideas (profiling cryptocurrency influencers in social media and trigger detection). As with the previous editions, PAN invites software submissions rather than run submissions; more than 400 pieces of software have been submitted from PAN’12 through PAN’22 combined, with recent evaluations running on the TIRA experimentation platform. This proposal briefly outlines our goals for PAN as a lab and our contributions proposed for PAN’23.