Data
To get the dataset, please send the scanned version of the duly filled-in "organizational-access form" (available at this link) to the email address aisome.fire2023@gmail.com. You can participate as an individual or as a team from one or more institutes. You can use a personal or institutional mail ID in the "organizational-access form". In the "organizational-access form" an individual researcher or organization can sign. Please mention "AISoMe FIRE 2023" somewhere in the form. Please mention the following in your mail: name of the team; and the name, affiliation and email id of each participant. There could be maximum 4 participant in a team. Single copy "organizational-access form" should be filled up team-wise.
Dataset Description of the Task
Training dataset: We will be providing a training dataset from our prior work “CAVES: A dataset to facilitate explainable classification and summarization of concerns towards COVID vaccines” (updated version on Arxiv: https://arxiv.org/abs/2204.13746, original publication (in Proc. SIGIR'22) on ACM portal: https://dl.acm.org/doi/abs/10.1145/3477495.3531745). The dataset contains 9,921 anti-vaccine tweets about COVID vaccines (that were posted during 2020-21), where each tweet has been labelled with one or more of the above labels by human annotators.
Test set: We plan to get around 500 more tweets annotated with the above labels, and release it as the test set, providing only the tweet IDs and the tweet texts for them. The tweets in the test set will be not only about COVID vaccines, but also other types of vaccines (e.g., MMR vaccine, Flu vaccine).