Using NLP To Reduce Wealth Flowing Into Local Human Trafficking Economies4:55pm - 5:20pm on Friday, October 4 in PennTop North
With $150 billion generated in profit a year, human trafficking is an illicit business model that affects children and adults in every community in the United States. Any strategy to reduce the prevalence of the crime must include reducing the amount of profit. Working with the New York Police Department, I built a prototype using NLP to target the buyers contributing the wealth to New York’s human trafficking economy. That prototype went on to serve nine jurisdictions across the United States and reach tens of thousands of buyers.
This talk explores the development of the chatbot engaging with buyers every hour of every day across the country to reduce the economic incentives behind human trafficking. We’ll explore scoping the domain of an NLP problem, designing the right machine learning model to address it, and creative ways to collect sufficient data to train an effective model.