Natural Language Processing in the Social Sector — Supplemental Materials

Pre-Processing, Parts of Speech, and Named Entities

The slides below illustrate basic pre-processing steps and two of the NLP techniques covered in the article: Part-of-Speech Tagging and Named Entity Recognition.

Sample code

More demos

Classification — “We’ve seen text like this before.”
Term Frequencies — “Let’s rank words by importance.”
Text Summarization — “What’s the TL;DR version of this text?”
Topic Modeling — “I think I really have five categories of text here.”

Acknowledgement

This site was created by DataKind. We would like to thank our contributors, who volunteered countless hours to develop the material on this site (including the sample code) in support of making NLP accessible to the social sector.

Sarah Eltinge is a data engineer based in Providence, RI who uses data to improve health and healthcare. In her free time, she enjoys knitting, biking, vegetarian cooking, and doting on her rescue cat.
Matthew Harris is a data science manager working for a fintech company in New York City. He plays guitar badly and fosters dogs for Badass Brooklyn animal rescue. All his furniture has been slightly chewed.
Jared McDonald is a software engineer in Brooklyn, New York. These days, his free time is mostly occupied by his rescue pup, Ginger, but he definitely had other interests before adopting her.
John Winter is a Data Scientist at Capital One in New York City. He has not rescued any animals, but volunteers at his church to make up for it. He enjoys yoga and meditation.

This site is open source. Improve this page.