Content Analysis of Cannabis Discourses on Twitter/X in the U.S.

Introduction: With the legalization of both medical and recreational cannabis use in many U.S. states, this study aims to explore public perceptions and discussions about cannabis on social media in the U.S. Methods: Twitter (now rebranded as X) data on cannabis were collected between February 2022 and February 2023 using the Twitter/X streaming Application Programming Interface. To assess the attitude of tweets toward cannabis and to determine whether Twitter/X users were cannabis users, human-guided deep-learning models called bidirectional encoder representations from transformers were used. The sex and age of users were inferred using a deep-learning facial recognition algorithm (DeepFace). The Latent Dirichlet Allocation topic model was used to comprehend the discussed topics. Results: Among 2,865,562 unique noncommercial cannabis tweets from the U.S., 648,018 tweets (22.62%) had a positive attitude toward cannabis, 234,202 (8.17%) had a negative attitude, and 1,983,342 (69.21%) had a neutral attitude. Among 821,451 unique Twitter/X users, 348,795 (42.46%) were potential cannabis users. The U.S. states allowing recreational cannabis use had 12.19 Twitter/X and cannabis users per 10,000 population, compared to 7.22 users in states without it; however, the difference was not statistically significant ( Conclusions: This study provides a detailed overview of public perceptions of cannabis in the U.S., aiding policymakers and public health authorities in developing effective regulatory policies about cannabis.