News for hackers. Stuff that matters.

A comment on Extracting financial disclosure and police reports with OpenAI Structured Output.

Re:Extracting financial disclosure and police reports with O... by Zaheer on Monday October 14, 2024 @12:40PM (#41839199)
Made a small project to help extract structure from documents (pdf,jpg,etc -> JSON or CSV): https://datasqueeze.ai/

There's 10 free pages to extract if anyone wants to give it a try. I've found that just sending a pdf to models doesn't extract it properly especially with longer documents. Have tried to incorporate all best practices into this tool. It's a pet project for now. Lmk if you find it helpful!

[ Parent | on HN ]
Re:Extracting financial disclosure and police reports with O... by hackernewds on Monday October 14, 2024 @02:09PM (#41840184)
Is this simply the OCR bits to feed to openai structured output?
[ Parent | on HN ]
Re:Extracting financial disclosure and police reports with O... by matchagaucho on Monday October 14, 2024 @01:09PM (#41839525)
Similarly I've found old-school OCR is needed for more reliability.
[ Parent | on HN ]
Re:Extracting financial disclosure and police reports with O... by bagels on Monday October 14, 2024 @01:39PM (#41839832)
Combining google's ocr with llm gives OCR superpowers. Tell the llm the text is from an ocr and ask it to correct it.
[ Parent | on HN ]
Re:Extracting financial disclosure and police reports with O... by saturn8601 on Monday October 14, 2024 @04:45PM (#41841797)
That sounds like it could be very dangerous when the LLM gets it wrong...
[ Parent | on HN ]
Re:Extracting financial disclosure and police reports with O... by bagels on Tuesday October 15, 2024 @01:25AM (#41845212)
Depends what you're using it for. If you're relying on OCR, you've already got to accept some amount of error.
[ Parent | on HN ]
Re:Extracting financial disclosure and police reports with O... by MarkMarine on Monday October 14, 2024 @05:02PM (#41841968)
I've been using this to OCR some photos I took of books and it's remarkable at it. My first pass was just a loop where I'd OCR, feed the text to the model and ask it to normalize into a schema but I found out just sending the image to the model and asking it to OCR and turn it into the shape of data I wanted was so much more accurate.
[ Parent | on HN ]
Re:Extracting financial disclosure and police reports with O... by artisandip7 on Monday October 14, 2024 @01:01PM (#41839447)
tried it works great, ty!
[ Parent | on HN ]