Automated Data Entry Unstructured Documents that Can Learn

1
Automated Data Entry Unstructured Documents that
Can Learn
2

• What if I said we have a solution that can
  automatically capture, classify and index
  documents automatically?
• Would you be interested?

3
What is advanced capture?

• Each and every capture process involving paper
  has the same problem.
• How to reduce the cost of scanning and data
  entry?
• Best-of-breed unstructured capture solutions that
  front-end SmartSeach
• Paradatec Prosar-Aida
• Kofax Transformations
• Captaris DoKuStar
• Advanced Capture provides
• Intelligent collation of documents WITHOUT
  separator sheets
• Auto classification
• Non-templated, unstructured, intelligent data
  extraction
• Quality assurance module for validating
  low-confidence data reads

4
Advanced capture applications

• AP invoice
• Process more than 3,000 invoices per month
• Companies that work with many different vendors
• AP GL account coding
• Invoice approvals
• XML integration with most any AP system
• Sales orders
• Process more than 3,000 sales orders per month
• Current process feeds order entry system
• Majority of sales orders come from fax, email or
  mail
• Explanation of Benefits (EOB)
• Hospitals, large specialty practices and 3rd
  party healthcare billing companies
• EOBs, super bills received via fax, mail or
  courier
• Majority of EOBs that are paper-based
• Current process feeds a patient billing system

5
Advanced capture applications

• Any set of unstructured documents that may have
  various layouts
• Tax forms
• Mortgage documents
• Real estate title documents
• Medical records
• Utility bills
• State labor documents
• Transcripts

6
Paradatec.

• Why Paradatec?
• More reasonable price point
• Simplistic pricing model
• Fastest OCR on the market
• NO OEM toolkits
• Been doing unstructured capture since 1988
• Standard output to XML
• Connectors into leading capture vendors like
  Kofax and Captiva
• Built-in rule trees for AP invoice and EOB
  processing to shorten PS effort
• What are the components?
• PROSAR - OCR/ICR, artificial intelligence,
  free-form text recognition engine
• PROKEY - correction and manual verification of
  data which could not or only partly be recognized
  by any of the two recognition programs.

7
How Does PROSAIR-AIDA Work?

• Step 1 Scan paper No separator sheets, mixed
  set
• Step 2 Document classification
• Stage 1 Quality assurance that the Paradatec
  engine properly classified the document
• Step 3 Data Extraction
• Stage 2 Quality assurance that data extracted is
  correct
• Step 4 Output of data and images

8
Stage 1 - classification

• Classification by keyword search from OCR output
• Spatial relationships between keywords is VIP
• Hierarchical classification allows thousands of
  types
• Can define as many clues as needed

9
Stage 2 - extraction

• Label-value pair logic is used to find data
• In this case we find all synonyms of invoice
  date
• Support for multi-page tables is inherent
• Output includes a single multi-page table

10
Provable Results

• Prepare a representative sample of images
• Customer keys data from images (perfect data)
• PROSAR AIDA automatically generates data from
  the same images (initially imperfect data)
• ResComp tool compares the two data sets
• Allows implementers to focus on exceptions and
  specific areas of improvement

11

• What are some real-life examples of
• PROSAR-AIDA in action?

12
Follett Higher Education Group

• Company Publishing
• Application Accounts Payable
• Captivas InputAccel document imaging platform
• Paradatecs PROSAR-AIDA OCR ICR technology
• Interfacing to a Lawson payment system
• Volume 30,000 pages per day 
• Capture 3 scanners, connecting to 3 PROSAR-AIDA
  systems via Captiva InputAccel. 
• Document Types Eleven (11)  

13
Akron Cardiology

• Company Healthcare, cardiology practice in Ohio
• Application Explanation of Benefits (EOB)
• Scan is performed by Kofax Ascent Capture.
• Data extraction and validation are performed by
  PROSAR-AIDA and PROKEY respectively.
• Volume 5,000 EOBs per month or 50,000
  pages/month
• Capture Kofax Ascent Capture
• Document Types Three EOBs, HCFA and UB

14
Akron Cardiology

• Challenge
• Extract data from potentially many hundreds of
  different EOB layouts
• Find a solution with a proven track record
• Eliminate the need for paper retention
• Increase the amount of data captured to include
  data needed to recover revenue (e.g. zero amounts
  paid procedures, paid amounts not matching
  contract, inappropriate denial codes)
• Results
• Data automation rates in excess of 75 across all
  insurance company layouts
• Eliminated the majority of manual data entry
  work 
• Seamless interface into document imaging system
• Provision of data needed in order to generate
  exception reports (e.g. procedures where "paid"
  amount is zero ....)
• Increased accuracy of data
• More data can be provided for customer service
  databases than could have been possible via
  manual data entry

15
Americas car-mart

• Company Automotive retailer with 91 locationss
• Application Accounts Payable and Sales
• Six (6) AP clerks sorted the invoices,
• Manually keyed them into the accounting system
  for payment,
• Filed them in cabinets
• Manual process was slow and prone to lost
  documents.
• Early payment discounts were often not realized.
• Company was not in a position to negotiate
  enterprise level agreements or price breaks
• Volume 100,000 per month
• Capture Kofax Capio
• Document Types Sales documents and invoices

16
DEMONSTRATION
Add approval routing
Augment with auto check run
17

• For more information
• Brent Wesler
• VP of Business Development
• Office 1 203-789-8791 ext.119
• Mobile [phone]
• bwesler_at_square-9.com