Active search of connections for case building and combating human trafficking

• NCJ Number: 306184
• Author(s): Reihaneh Rabbany; David Bayani; Artur Dubrawski
• Date Published: 2018
• Length: 10 pages
• Grant Number(s): 2013-IJ-CX-K007
• Agencies: NIJ-Sponsored
• Publication Type: Research (Applied/Empirical)
• Award: Adaptive Expert System that Learns to Detect and Track Patterns of Crime in Internet Advertisements

Annotation

The reported experiments focused on case building for combating human trafficking, where the investigator follows leads to expose organized activities, i.e., different escort advertisements that are connected and possibly orchestrated.

Abstract

The authors’ “RedThread,” is an efficient solution for inferring related and relevant nodes while incorporating the user's feedback to guide the inference. The results of RedThread are interpretable, as they explain how the results are connected to the initial lead. The authors experimentally show that RedThread learns the importance of the different types and different pieces of evidence, while the former could be transferred between cases. (Published abstract provided)