JOURNAL ARTICLE

Detecting Human Trafficking: Automated Classification of Online Customer Reviews of Massage Businesses.

  • Published In: Manufacturing & Service Operations Management (M&SOM) (INFORMS), 2023, v. 25, n. 3. P. 1051 1 of 3

  • Database: Business Source Ultimate 2 of 3

  • Authored By: Li, Ruoting; Tobey, Margaret; Mayorga, Maria E.; Caltagirone, Sherrie; Özaltın, Osman Y. 3 of 3

Abstract

The article focuses on developing natural language processing (NLP) methods to detect illicit massage businesses (IMBs) involved in human trafficking by analyzing online customer reviews, particularly from open platforms like Yelp.com. It presents a novel approach combining lexicon-based and embedding-based classification models, including those using BERT and Doc2Vec embeddings, to identify reviews indicating commercial sex or trafficking risk factors. The study creates a labeled dataset of Yelp reviews with expert input, develops a specialized lexicon of trafficking-related terms, and applies data augmentation techniques to address class imbalance. Ensemble models combining lexicon and embedding approaches achieve improved detection performance, offering law enforcement and victim service organizations tools to prioritize investigations and outreach. The research highlights the potential for expanding these methods to other online platforms and related industries that may serve as fronts for trafficking.

Additional Information

  • Source:Manufacturing & Service Operations Management (M&SOM) (INFORMS). 2023/05, Vol. 25, Issue 3, p1051
  • Document Type:Article
  • Subject Area:Law
  • Publication Date:2023
  • ISSN:1523-4614
  • DOI:10.1287/msom.2023.1196
  • Accession Number:163811512
  • Copyright Statement:Copyright of Manufacturing & Service Operations Management (M&SOM) (INFORMS) is the property of INFORMS: Institute for Operations Research & the Management Sciences and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Looking to go deeper into this topic? Look for more articles on EBSCOhost.