This project addresses the use of machine learning techniques on massive on-line information for identifying patterns useful in law enforcement practice and public policy. The quantity and complexity of public data make it impractical for manual browsing and tracking by investigative forces seeking to identify illegal activities. For example, there is a need to automate the process of detection of informative patterns in human trafficking, and especially sex trafficking. Ubiquity of the Internet provides the perpetrators, who take advantage of trafficking victims, with the ability to solicit services on advertisement web sites. This data is publicly available and contains information potentially useful to law enforcement investigators to understand crime patterns and to track the perpetrators. This team proposes to study the commercialization opportunity of a prototype analytic tool which they developed for the task. This technology may save time and increasing law enforcement productivity and effectiveness by providing new leads that would otherwise be missed.

This project provides a framework for demonstrating practical utility of machine learning research previously funded by NSF in a well-focused context of its societally important application. The proposed innovation has the potential to revolutionize the investigative process in the US at local, state, and federal levels. The vast amount of data online is currently underutilized. This project seeks to remedy this for general benefit of society. It would improve efficiency of law enforcement practice at all levels (local, state, federal), facilitate cross-agency collaboration, as well as enhance public policy studies. The proposed innovation will also have a broader application in leveraging public sources of data for detection and mapping of patterns of other illicit behaviors (e.g. sale of stolen and counterfeit goods) and non-illicit behaviors (e.g. economic activity, job seeking, lifestyle, and public health).

Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Carnegie-Mellon University
United States
Zip Code