This guide will walk you through Rossum’s app. You will learn how to:
Create an account
To create an account, head to Rossum registration page. In the first step, you’ll need to select your region to define the list of Pre-trained fields that will be extracted (Please note: this can be changed later).
Note: Although Rossum currently supports pre-trained fields only for Invoice processing, the technology is document agnostic and can extract data from any structured document including receipts, purchase orders, shipping documents, etc.
Next, fill out your personal details including First name and Surname, Company, and Phone number.
Then you can choose the business email address you want to associate with the account and a secure password. Your trial account is free for <300 documents per month and is fully configurable.
Note: Automated learning of your documents is available only in higher editions.
Welcome to the app
Once you log in, you will arrive at the Main screen.
On the left-hand side, there is a list of Queues. Each queue is an organizational unit that typically represents a particular document type that needs to be processed. The left panel is fully customizable, so you can add/group/remove queues as needed.
Read more: How to configure fields for data extraction
In the center of the screen, you can see the Document list with a pre-uploaded sample invoice.
Upload your documents
Now, you can start uploading your documents by selecting the “Import” button or by sending your documents to the specified email address that is given (this can also be done via API as well). You can import PDFs or scanned images of your documents.
Click “Done” and the AI will automatically begin extracting the data fields. It typically takes between 30-60 seconds per document for the extraction.
After processing, the documents can be found in the “To review” tab in case human review is needed.
Validate the data
When you select a document for review, it will take you to the Validation screen. In the left panel, you can see the predefined list of fields which have been extracted. The “eye icon” suggests to manually review the field, due to the lower confidence of the automated extraction.
Press Enter to check the fields one by one. Fields marked with a red flag must be reviewed and corrected, otherwise, you won’t be able to export the data. If you realize that data field was not extracted correctly, you can make adjustments by:
- clicking and pointing
- dragging the rectangle
- typing the text directly
Export the data
After the validation, click “Confirm” the document will appear in the Exported tab on the main dashboard. There, you can select to download data in the multiple formats (CSV, JSON, XML, via API)