TRAFFICVIS: Visualizing Organized Activity and Spatio-Temporal Patterns for Detecting and Labeling Human Trafficking

Catalina Vajiac, Duen Horng Chau, Andreas Olligschlaeger, Rebecca Mackenzie, Pratheeksha Nair, Meng-Chieh Lee, Yifei Li, Namyong Park, Reihaneh Rabbany, Christos Faloutsos

View presentation: 2022-10-19T14:12:00Z GMT-0600 Change your timezone on the schedule page
Exemplar figure, described by caption below
Analyzing online escort ads using TrafficVis: we show one meta-cluster, i.e. micro (text) clusters connected using metadata, on real data. Some text blurred for privacy. 1. Human trafficking domain expert uses Micro-cluster panel to drill down to specific micro-cluster data and associated ads. 2-3. Expert uses Timeline panel and Map panel to investigate metadata, noticing inconsistent posting time and regional geographic spread, ruling out spam and scam. 4. Expert uses Text panel to quickly find telling signals; differences between ads in a micro-cluster are highlighted. 5. Finally, the expert confidently labels the meta-cluster for each modus operandi (M.O.), deciding on benign (at-will sex worker), with a small chance of trafficking.

Prerecorded Talk

The live footage of the talk, including the Q&A, can be viewed on the session page, Decision Making and Reasoning.

Fast forward

Law enforcement and domain experts can detect human trafficking (HT) in online escort websites by analyzing suspicious clusters of connected ads. How can we explain clustering results intuitively and interactively, visualizing potential evidence for experts to analyze? We present TrafficVis, the first interface for cluster-level HT detection and labeling. Developed through months of participatory design with domain experts, TrafficVis provides coordinated views in conjunction with carefully chosen backend algorithms to effectively show spatio-temporal and text patterns to a wide variety of anti-HT stakeholders. We build upon state-of-the-art text clustering algorithms by incorporating shared metadata as a signal of connected and possibly suspicious activity, then visualize the results. Domain experts can use TrafficVis to label clusters as HT, or other, suspicious, but non-HT activity such as spam and scam, quickly creating labeled datasets to enable further HT research. Through domain expert feedback and a usage scenario, we demonstrate TrafficVis's efficacy. The feedback was overwhelmingly positive, with repeated high praises for the usability and explainability of our tool, the latter being vital for indicting possible criminals.