Is anybody using OCR (optical character reader) in a workflow?

[Philip]

We have scanned PDFs coming in as a data source. Is anybody using OCR in a workflow? If so, which one and how has it worked for you?

[patrick_mcauliffe · Accepted solution]

@Philip, Check out this post: https://community.alteryx.com/t5/Advanced-Analytics/Image-Face-Recognition-Using-Microsoft-Cognitive...

One of the other API's included in the Microsoft Cognitive Services is an OCR API.  It follows the same type of API call as the Face API described in the post.

I've found it to be fairly accurate.

[Philip]

Thanks, Patrick! This looks like a great possibility.

[Philip]

I've been told we'll be getting Azure here, so this solution will work out great. Thanks for sharing it, @patrick_mcauliffe!

[khu]

Hi, I'm also looking for OCR for image pdf, Have you created the workflow using the API?