Apologies for cross-posting.
----------------------------------------
We kindly invite you to participate at the Pan-2023 Task 2 - “Profiling Cryptocurrency Influencers with Few-Shot Learning”. This task is being held as part of CLEF 2023, and all participating teams will be able to publish their system description paper at the CLEF proceedings. This shared task focuses on the author profiling of cryptocurrency influencers in social media from a low-resource perspective, that is, with little training data. Moreover, we propose to profile types of influencers also using a low-resource setting.
Specifically, we focus on English Twitter posts for three different sub-tasks: • Low-resource influencer profiling (subtask-1): profile authors according to their degree of influence (null, nano, micro, macro, mega). • Low-resource influencer interest profiling (subtask-2): profile authors according to their main interests or areas of influence (technical information, price update, trading matters, gaming, other). • Low-resource influencer intent profiling (subtask-3): profile authors according to the intent of their messages (subjective opinion, financial information, advertising, announcement).
Important Links • Task Website - https://pan.webis.de/clef23/pan23-web/author-profiling.html • Dataset site - https://zenodo.org/record/7701748#.ZBxxrnbMKiM
Important Dates • February 20, 2023: Training data ready • May 10, 2023: Early bird software submission phase (optional) • May 29, 2023: Software submission deadline • June 05, 2023: Participant paper submission • September 18-21, 2023: Conference
Task organizers • Francisco Rangel (Symanto) • Mara Chinea-Rios (Symanto) - Contact Email: mara.chinea@symanto.com • Marc Franco-Salvador (Symanto) • Paolo Rosso (Universidad Politécnica de Valencia)
Please reach out to the organizers at crypto-influencers-pan-organizers@googlegroups.com, or join the Slack workspace (https://pan2023profil-0q48349.slack.com) to connect with the other participants and organizers.