ChatGPT to extract data from PDFs

Screen shot of a page with a table split across a two column layout. The table has four actual columns.

PDF of a table with Social Security tax info

Just a quick note on a useful application of ChatGPT. You can use it to extract just the subset of the data you want from a table in a pdf document, a really nice time saver.

I wanted the year, max. earnings, and OASDI tax rate for the years 1982 through 2022 from the pdf page here: taxpolicycenter.org/sites/defa. A real pain to cut and paste into excel given the double column and then convert to numbers. So I gave ChatGPT the URL and listed the years and columns I wanted. And I got back a nice table with exactly what I asked for. Nice time saver!

Leave a Reply

Your email address will not be published. Required fields are marked *

The maximum upload file size: 512 MB. You can upload: image, audio, video, document, spreadsheet, text. Links to YouTube, Facebook, Twitter and other services inserted in the comment text will be automatically embedded. Drop files here