0% found this document useful (0 votes)
22 views28 pages

Document Understanding Webinar

The document discusses UiPath's Document Understanding, which integrates AI and RPA to automate document processing, leading to significant cost reductions and efficiency improvements. It outlines the framework for document processing, including digitization, classification, extraction, and validation, and highlights the benefits of automating these tasks. A live demo and case study on invoice processing illustrate the practical applications and outcomes of implementing this technology.

Uploaded by

thrinath reddy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
22 views28 pages

Document Understanding Webinar

The document discusses UiPath's Document Understanding, which integrates AI and RPA to automate document processing, leading to significant cost reductions and efficiency improvements. It outlines the framework for document processing, including digitization, classification, extraction, and validation, and highlights the benefits of automating these tasks. A live demo and case study on invoice processing illustrate the practical applications and outcomes of implementing this technology.

Uploaded by

thrinath reddy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 28

Fast and Accurate Document

Processing with AI-enhanced


Document Understanding
Meet your speakers

Ioana Gligan Robert Love Razvan Nutica Sophia Sembai


Product Manager, Presales Technical Consultant, Automation Center of Product Marketing Manager,
Document Understanding Artificial Intelligence Excellence (CoE) Head Artificial Intelligence

2
Agenda for today

1. Introduction to document understanding


2. UiPath Document Understanding as a way to process documents intelligently
3. Live demo
4. UiPath Automation CoE case
5. Q&A session
Introduction to
document understanding
Document understanding emerges at the intersection
of document processing, AI, and RPA

Not
OCR

Not
Computer
Vision

5
Why to automate document processing?

35% 52% 17% 40%


growth in
cost reduction decrease in errors reduction in time
employee
compared to that mitigates the employees spend
productivity or
manual document risk of rework and on document
customer
processing related losses processing
satisfaction

6
“Respondents across all industries said content intelligence
technologies enabled several key corporate initiatives,
including employee engagement, customer engagement,
work transformation, and overall digital transformation”

Holly Muscolino
International Data Corporation (IDC)

7
UiPath
Document Understanding as
a way to process documents
intelligently
Get documents processed intelligently with
AI-enhanced UiPath Document Understanding

9
Every company in the world processes documents

Human
Financial Services
General Resources Manufacturing Public Sector Healthcare
& Insurance
(HR)
• Invoice • Accounts Payable • Employee • Sales order • Immigration • Medical
• Receipt & Accounts onboarding processing application forms
• Purchase Receivable • Resume • Customer • School • Medical bills
Order • IRS form screening parts application • Health
• Utility bill • Loan application • HR records request • Passport records
• Bill of • Mortgage processing • Remittance management • Drug
Landing processing processing application prescriptions
• Passport • Account opening
• License and customer
onboarding
• Claims
processing
• Vendor
onboarding
• Compliance-
related processes

10
It’s time to automate!

Accelerated Higher Better customer Happier


productivity efficiency experience employees
Automate highly-manual Save time and costs
document processing spent on paperwork with Mitigate the risk of errors Help employees escape
tasks to accelerate easy to deploy and and decrease the from the mundane chores
productivity rates maintain automations response time to deliver and focus on higher-value
better customer tasks
experience

11
Document Understanding Framework

Load Digitize Classify Extract


taxonomy Export
to obtain and split the information
to define doc machine- from the the extracted
files into
types and readable text document information for
using OCR if document using different
fields for further usage
extraction necessary types methods

Validate
if needed - a human can help
confirm the classification or data
extraction results
12
Document Understanding Framework

Load Digitize Classify Extract Export


taxonomy to obtain information
and split the the extracted
to define doc machine- from the
files into information
types and readable text document
using OCR if document using different for further
fields for
extraction necessary types methods usage

Validate
if needed - a human can help
confirm the classification or data
extraction results
13
Digitize text in the documents using OCR
Move-For-You Co.
We move so you don’t need to move
PO: NP74006735
1 February 2020
PAYABLE WITHIN 15 DAYS OF RECEIPT
20800 ALMADEN AVE, SUITE 404
SAN JOSE, CA 95120-0520
T: +1 425 555 9876
F: +1 425 555 3456
E: [email protected]
www.moveforyou-co.com
Bill To:
Tony Tzeng
12345 Mango Lane
Seattle, WA 98108
INVOICE DETAILS
Packing services
Storage fees (1 month)
House move (white-glove service)
Vehicle storage and transport
Sales tax 10%
Total Fee including Tax
FEE
$1,282.00
$1,884.00
$5,320.00
$5,186.00
$1,367.20
$15,039.20
Methods of payment
Personal Check: Move-For-You LLC
Wire Transfer: BigBank Co., Account 123456789-0987ABC
Invoice No: 456200-TZE1 14
Classify and split the documents
______
Purchase
______
Invoice
order
______
______ ______
______
______ ______
______
______
Multi-page
document

______
Receipt ______
______
______ ______
Unrelated
______
______ ______
documents
______
______
______ ______
_____
______
_____

15
Validate classification of the documents

16
Extract information from the document

Rule-based Model-based Custom

RegEx-Based Form Intelligent Form Machine Learning Bring your own


Extractor Extractor Extractor Extractor extractor

Hybrid approach

A combination of both –
rule-based and model-based extractors

17
Rule-based or template-based approach

Relies on rules (like regular Processes fixed in format Ensures high accuracy for
expressions) and templates structured data already known documents

18
Machine learning (ML) models
as a template-less approach

Pre-trained ML models Bring your own model Model retraining


working as-is
Create new custom model AI Fabric
Invoices (+ for India and Australia*)
Receipts Third-party models Validation Station**
Purchase Orders*
Utility Bills*

* in public preview, will become generally available with the 20.10 release 19
** will be released with the 20.10 release
Validate extracted data

20
Flexible
Process various types and formats of documents, even unstructured ones or varying layouts.

Intelligent
Mix & match different extractors and retrain ML models to teach robots accurately understand
and act upon documents.

End-to-end
Seamlessly automate high-volume document processing with end-to-end automation.

Open & extensible


Bring your own or third-party components and use them within Document Understanding.

21
Roadmap for 2020

More pre-trained Retraining loop Increased language


templates and from human support for
models validation handwriting and
documents ​

22
UiPath
Document Understanding
live demo
Example: Document Processing

Start Document Understanding Digitized Documents

Multiple Shared file UiPath OCR engines


document folder Robot
types

______
______ Expense
______ Forms
Intelligent Financial
______
Reports
______
______
_____
Classifier and ______

Splitter
Attended Robot &
Export ______
Validation Station ______
______ ______
validated data ______
______
______
______
______ ______
Extracted Classified & split _____ Invoices
______
______
Misc
______
______
_____ _____
document details documents
______ ______

Rule-based
Action Center & & model-based
Validation Station extractors

24
UiPath Automation CoE case
Use case: Invoice processing
Check for new e-
Extract e-mail body Prepare extracted
mails with attached
and attachment(s) data set(s)
invoices
Supplier Email Unattended
Robot

Document Unattended
Understanding Robot
Create and
Action Center​ assign tasks
Orchestration
process

Validated
Unattended data set
Robot
Tasks management dashboard Validation station (task view)

Open bill
in Coupa 26
Coupa API layer
First business outcomes

1,350 hours
80% of time savings

$27,000
saved costs spent on manual invoice processing

8,100 invoices
processed by robots

27
Q & A session.
Thank you!

uipath.com/document-understanding

© 2005–2020 UiPath. All rights reserved.

You might also like