This research develops an innovative NLP framework that leverages large language models to detect drug use patterns and overdose symptoms from social media posts.

• Combines human expertise with AI capabilities through a hybrid annotation approach
• Enables real-time monitoring of substance use trends and potential overdose risks
• Offers a scalable alternative to traditional research methods that face limitations in capturing current drug use patterns
• Creates opportunities for earlier public health interventions by identifying emerging substance abuse trends

For medical professionals, this technology represents a breakthrough in population-level surveillance of drug use behaviors, potentially enabling more timely and targeted intervention strategies to address overdose risks before they escalate to crisis levels.

Leveraging Large Language Models for Multi-Class and Multi-Label Detection of Drug Use and Overdose Symptoms on Social Media