Hidden in Plain Sight: A Machine Learning Approach for Detecting Prostitution Activity in Phoenix, Arizona

Authors: Edward Helderop*, , Jessica Huff, Arizona State University, Fred Morstatter, University of Southern California, Anthony Grubesic, Arizona State University
Topics: Urban Geography, Applied Geography, Quantitative Methods
Keywords: Natural language processing, machine learning, prostitution, spatial analysis
Session Type: Poster
Day: 4/13/2018
Start / End Time: 1:20 PM / 3:00 PM
Room: Napoleon Foyer/Common St. Corridor, Sheraton, 3rd Floor
Presentation File: No File Uploaded

Prostitution has been a topic of study for decades, yet many questions remain about where prostitution occurs. Difficulty in identifying prostitution activity is often attributed to the hidden and seemingly victimless nature of the crime. Despite numerous challenges associated with policing street prostitution, these encounters become more difficult to identify when they take place indoors, especially in locations away from public view, such as hotels. The purpose of this paper is to develop a strategy for identifying hotel facilities and surrounding areas that may be experiencing elevated levels of prostitution activity through the use of high-volume, user-generated data, namely hotel reviews written by guests and posted to Travelocity.com. A unique synthesis of methods, including data mining, natural language processing, machine learning and basic spatial analysis are combined to identify facilities that may require additional law enforcement resources and/or social/health service outreach. Prostitution hotspots are identified within the city of Phoenix, Arizona and policy implications are discussed.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login