Abstract: Illegal drug transactions have been a challenging issue in the broader Southern California area for decades. In recent years, such transactions have shifted to darknet forums, which enforce the anonymity of traders through encrypted communication and increase the difficulty of drug control. In this work, we aim to leverage machine learning and data science techniques to address this public health concern in Southern California and across the United States. Specifically, we emphasize the importance of exploring these forums and highlight the challenges of research in this field. In addition, we introduce several widely used darknet datasets and present several research problems including stylometry-based authorship attribution and trust signaling strategies on darknet platforms.