Resources · Archive

Patent Data & Parsing Code

In 2013, I wrote Python code to download and clean information pertaining to U.S. patents issued since 1980. This preserved archive contains sample flat files, the original Python and Stata parsing code, and documentation.

Sample data

Flat files

Small files containing 1,000 records each.

Main bibliographic file Inventors Assignees Citations to other patents Citations to non-patent literature Field of search Parent-child linkages Non-primary classes

Code & documentation

Replicate and extend

The original scripts and notes on extending the files to later years.

Python and Stata parsing code README

The original full-scale flat files are preserved in the Bell Labs backup. Because they total more than two gigabytes, they will be restored as separate bulk downloads rather than bundled into every website release.