A Machine Learning Approach for the Detection and Characterization of Illicit Drug Dealers on Instagram

 

Social media has now become ubiquitous, but this growth has also made it a veritable toolbox for drug trafficking. Despite increased scrutiny by the media, regulators and policymakers, Instagram continues to be a popular platform for drug dealers, in violation of federal law. The aim of this study was to develop and evaluate a machine learning approach to find Instagram posts related to illegal drugs.

Using a combination of Web scraping and deep learning, researchers were able to detect drug trafficking with high accuracy. From the 12,857 posts collected, they detected 1,228 drug trafficking posts consisting of 267 unique users. They then used cross-validation to evaluate the 4 models, with their deep learning model reaching 95% on F1 score and performing better than the other 3 models.

Learn more about their important work towards reducing drug trafficking posts on social media platforms by downloading the report.

julia.guerra@cintoc.org

Authors:
Jiawei Li
Qing Xu
Neal Shah
Tim K Mackey