Social Media Text Mining Framework for Drug Abuse: Development and Validation Study With an Opioid Crisis Case...

Area of Science:

• Social media analytics
• Public health informatics
• Computational social science

Background:

• Social media offers valuable data for understanding disease trends and patient behaviors.
• Analyzing social media for drug abuse insights faces challenges in topic deduction and data quality.
• A structured approach is needed to overcome these limitations in social media data analysis.

Purpose of the Study:

• To develop and evaluate a framework for mining and analyzing social media content related to drug abuse.
• The framework aims to mitigate challenges in topic deduction and data quality for social media data analytics.
• Enhancing the reliability and interpretability of social media data for drug abuse research.

Main Methods:

• Developed a drug abuse ontology to define keywords, categories, and characteristics.
• Utilized the Crimson Hexagon platform for data collection based on the ontology.
• Implemented data preprocessing and quality evaluation using a matrix for robust analysis.

Main Results:

• The framework was applied to the opioid epidemic, demonstrating its utility.
• Successfully identified public concerns and prominent discussion topics related to the opioid epidemic on social media.
• Showcased improved topic discovery in diverse and complex social media domains.

Conclusions:

• The proposed framework effectively addresses challenges in topic detection and data quality.
• Validated the framework's applicability in identifying public concerns and key discussions on social media regarding the opioid epidemic.
• Provides a robust method for analyzing social media data in the context of drug abuse and public health crises.