Build a Human Trafficking Dataset from Court Cases and News Articles¶

The goal is to build a dataset that gathers information on human trafficking from court cases and news articles. This project started from a series of conversations with Professor Yuliya Zabyelina, Eric Schles and Aida Shoydokova.

• 20171012 YZ suggests to RH that it would be useful to build a human trafficking dataset from court cases and news articles.
• 20171103 AS talks with RH and begins working on information extraction from court cases.
• 20171105 YZ, N, OC, AS, SA, RH meet to discuss ideas and coordinate work.
• 20171214 YZ, ES, AS, RH present their work in a workshop on Computational Approaches to Fight Human Trafficking Using Python, hosted at Civic Hall by CrossCompute, NYC PyLadies, Spatiotemporal Analysis for Community Health and Safety and sponsored by the Python Software Foundation.

Dataset Exploration¶

• Explore USA Department of Justice Court Case Press Releases

Text Preparation¶

• Extract Text from a Webpage
• Normalize Raw Text

Information Extraction¶

• Guess the Gender of a Name from the USA

Dataset Extraction¶

• Extract Human Trafficking Incidents from Court Cases using NLTK
• Extract Human Trafficking Incidents from Court Cases using spaCy

All of the exercises in this tutorial are also CrossCompute Tools. To learn how to format a Jupyter Notebook as a CrossCompute Tool, please see this example.