This video tutorial shows how to use UiPath Document Understanding and machine learning to extract data from many invoices. The invoices can even be in different formats.
We will extract data (invoice number, date, total, etc.) from a folder of invoices to an Excel file in this use case. We will also talk about the Present Validation Station, where users can verify the results in an attended robot.
You could also watch:
🔵 UiPath Advanced Use Cases - [ Ссылка ]-
🔵 UiPath REFramework Full Tutorial - [ Ссылка ]
🔵 Extract structured invoice data with Regex - [ Ссылка ]
0:00 Intro
We have multiple invoices in different formats that we want to automatically extract data from. The game plan is to use the Document Understanding package in UiPath.
0:53 Install package
We install 3 packages: Document Understanding, IntelligentOCR, and Omnipage.
2:10 Load Taxonomy
In the Taxonomy Manager, we specify what data fields we want to extract and load afterward.
4:36 Read our invoice
We are using the Digitize Document to read the invoice into a format and can work within UiPath.
6:12 Specify the OCR Engine
In this use case, we use the OmniPage OCR.
6:28 Classify Document Scope
Here we must classify, that is, sort the input (here our invoices, but we could have documents of other types). A keyword-based classifier is used. Remember to create an empty JSON.
9:44 Data Extraction Scope
We use a machine learning extractor. It is shown how to set up the extractor.
12:58 Present Validation Station
For attended use, we can use a validation step, in where the user can validate the extracted data.
14:05 Export Extraction Results
The data will be written to Microsoft Excel.
17:43 Scale the solution
To improve the solution from one invoice to all invoices in a folder, we modify our solution.
FOLLOW ME:
💼 LinkedIn: [ Ссылка ]
🐦 Twitter: [ Ссылка ]
📸 Instagram: [ Ссылка ]
👨 Facebook: [ Ссылка ]
👨👩👧👦 Discord: [ Ссылка ]
💌 Email Newsletter: [ Ссылка ]
#uipath #rpa #automation
