|
Activity Number:
|
181
- Late-Breaking Session 1: Human Trafficking Analytics in the Age of COVID-19
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, August 10, 2021 : 1:30 PM to 3:20 PM
|
|
Sponsor:
|
JSM Partner Societies
|
|
Abstract #319196
|
|
|
Title:
|
Identifying a minimal subset of online classified ads that are likely to be human trafficking recruiting attempts using text analytics
|
|
Author(s):
|
William Taylor* and Lauren Agrigento
|
|
Companies:
|
SAS and SAS
|
|
Keywords:
|
|
|
Abstract:
|
With the widespread availability of broadband internet across the globe, the human trafficking pandemic has evolved in its efforts to recruit vulnerable people into its grasp. Traffickers disguise predatory advertisements on classified ad websites as modeling, escort, masseuse and other gigs to attract young women looking for a quick paycheck. Identifying human traffickers on classified ad websites is impractical due to the massive number of ads being posted alongside them. How do we find the ads that are most likely to be written by human trafficking recruiters? One method to address this relatively new form of recruiting is to filter the ads based on their content. We build text topics from a set of ads that were verified by the Louisiana State Police to have been posted by human trafficking recruiters. We then use these text topics to categorize a larger set of ads that were scraped from BackPage.com as “recruiting." The categorization resulted in a 99.7% decrease in the number of ads that law enforcement officers would need to search through as part of continuing investigation.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2021 program
|
