What is CREAT?
Overview
The Census Research Exploration and Analysis Tool is a data
tool from the Center for Economic Studies (CES) at the US Census Bureau that uses natural
language processing and artificial intelligence tools to analyze, categorize, and sort the
economic research contained in the CES working paper series. The goal of this
project is to help CES researchers, managers, and other internal stakeholders explore
connections among existing research, form new collaborations, and separate research into
discrete topics. Working papers are sortable by author, tag, and keyword. For more
information, see the CREAT one-pager.
Keywords and Tags 
Keywords and tags are automatically extracted from the text of the working papers. Keywords are either one or two words, and are extracted to most closely match the research paper as a whole. Tags are institutions, datasets, and other proper nouns that occur with relative frequency among all working papers. Due to the automatic extraction, accuracy is not guaranteed.
Authors 
Authors are extracted from the paper and automatically matched using fuzzy matching to ensure that different forms of the same authors are consolidated, e.g., middle initial, title, etc.
Contact Us
If you have any questions or comments, contact us at ces.local.employment.dynamics@census.gov.