Autism Twitter Dataset

Autism spectrum disorder (ASD) is a developmental delay causing physical, cognitive, and behavioral changes and affecting millions of individuals. Because of the complexity in its symptom profiling and misdiagnosis, it is crucial to provide appropriate treatment on time and improve the efficacy of screening tools. Social media has turned out to be a useful means for real-time public health monitoring, giving rise to an unprecedented surge in user-generated content with millions of users sharing their thoughts, experiences, and health-related information. The digital footprint of an individual can be analyzed to study behavioral symptoms of ASD and other mental health disorders. This is a novel dataset consisting of 6,515,470 tweets in total, 3,137,952 of which are from 17,323 users self-identifying with ASD and 3,377,518 from 171,273 control group users. Along with the tweets the dataset also has supporting information such as posting dates, follower count, geographical location, and interaction metrics. The dataset can be used to study various aspects of social communication that differentiate autistic people from their neurotypical peers and offer multiple opportunities for future research using NLP analytics, such as using pre-trained large language models like BERT and GPT for text classification, topic modeling, and feature extraction.