Authentic data: vaccine position and stigma

All authentic data and annotations for vaccine position and stigma within the dataset for detecting vaccine positions and vaccine-related stigma on social media. The data is structured as follows:

Data ID: A unique identifier assigned to each record in the dataset. Identifiers beginning with ‘1-‘ are applied to authentic tweets; those beginning with ‘2-‘ refer to synthetic tweets.Authentic Tweet ID: A unique identifier assigned to each tweet by Twitter. As the paraphrases do not have unique Twitter IDs, this column instead presents the ID of the authentic tweet from which the paraphrase was generated.Tweet Paraphrase: The text presented in each paraphrase. In keeping with big data ethics, the text of the authentic tweets is not presented. Authentic tweets listed in the blended dataset present ‘N/A’ in this column as they are not paraphrases.Vaccine Position Class: Lists the annotation codes 0 (‘Pro-Vaxx’), 1 (‘Anti-Vaxx’), and 2 (‘Vaccine Hesitant’).Stigma Class: Lists the annotation codes 3 (‘Stigma’) and 4 (‘No Stigma’).Using the Twitter API and the tweet IDs, researchers can obtain access to the authentic tweets' text data provided that these have not been deleted or otherwise made unavailable.