Презентация на тему Kofax Transformation Modules 5. Module 2 – Product Overview

Module Overview
What is KTM?
Structured, semi-structured and unstructured documents
Classification, separation

and extraction
Kofax product differentiation
Licensing
Program Group components
Integration with Kofax Capture

Product Overview

is KTM?
KTM is a classification, separation and extraction “transformation”

product for Kofax Capture (a suite of custom module add-ons)

Although somewhat new to Kofax Capture, the product was introduced as Ascent Xtrata Pro 3.0. The product has now been rebranded as Kofax Transformation Modules, and this version is a 5.x release.

Supports classification and extraction of structured, semi-structured and unstructured documents.

Platform for delivering pre-configured, out-of-the-box solutions

Offers learn-by-example extraction as well as numerous other methods to automatically extract data from documents

Invoice classification by image and instructions (rules).

or Unstructured?
KTM – Includes processing modules, a configuration application

and several tools that work within Kofax Capture providing advanced document classification, separation, and extraction for structured, semi-structured and unstructured documents. It is optimized for invoice processing.

Structured

Unstructured

Semi-structured

– Distinguishing between different Document Types and Form Types.
Separation

– Creating boundaries between individual documents in a batch. Documents may be single-page or multi-page.
Extraction – The process of automatically lifting data from a document using rules or zones along with the output from a recognition engine (OCR, OMR, barcode, etc). This data usually is output to a database or document management application for later retrieval.

of KTM’s Features Include…
KCNS compatible
Enterprise Option (High Availability) compatible
Processes

image (.tif) files
2 variations
Invoice Projects
Standard Projects
Knowledge Bases
Generic and Specific Knowledge Bases
Country-specific
Based on “on-line learning”
Optional password protection
“Professional” add-on available
Includes advanced page classification and trainable document separation using the special “Project Planner” configuration tool
Other standard features
Database lookups
Regular expression support
Dictionary files
Table recognition
Scripting WinWrap Basic (similar to VBA)
Language-specific support in OCR (ABBYY Fine Reader, RecoStar, AEG, Kadmos)

KTM Server module
Classification and Separation
Extraction
Document Review module
Correction module: character-based

correction
Multiple instances of the KTM Validation module
KTM Verification module
Trainable Group Locator
SQL Server support for statistics export
Specific on-line learning for tables
Page-level and zonal image cleanup
Trainable Document Separation (Professional Version only)

Slide

● Module 2 - Product Overview

registration
Web Validation and Verification modules
Arabic OCR engine (right ?

left)
Upgraded ABBYY OCR engine
Integrated A2iA engine for cursive text and check recognition
Database (SQL and ODBC) lookup utility
Metadata-driven application setup
Vendor Locator
Invoice Evaluator
Country-specific settings and generic knowledge bases
Purchase order line item matching

Slide

● Module 2 - Product Overview

Support
Thin client enhancements
New Thin Client Correction
Enhanced Thin Client Validation

folder support
batch editing
multiple steps
Enhanced Document Review Module
Script support in Verification
Miscellaneous other features
New simplified licensing
Cursive Full-page Recognition

Page

● What´s new in Kofax Transformation Modules 5.0

Capture alone and with KTM Compared

numbers of fields extracted
5-Fields Extraction Base License
Unlimited Field

Extraction Base License
Both base licenses include
Layout and content classification
Fixed- and free-form extraction
Invoice header and amount extraction (GLs/IHL)
Keying modules (inc web clients)

Page

● What´s New in Kofax Capture 5.0

OCR engines (Kadmos)
Check Recognition
Cursive Field Recognition
Cursive Full-Page Recognition
Arabic OCR
Invoice

Add-On Pack (Vendor Locator & Invoice Evaluator)
Invoice Line Items Matching
Medical Claims Add-on Pack

Page

● What´s New in Kofax Capture 5.0

New in Kofax Capture 5.0

possible
Mixture of base license volumes for a single system
900k

total volume = 300k 5-field extraction + 600k unlimited field extraction
Different volumes for add-ons possible
30k Table Extraction + 120k Arabic OCR

Page

● What´s New in Kofax Capture 5.0

Group Components
NOTE: In this example, 5 instances of KTM

Validation have been installed. Project Planner (Professional Upgrade) has also been installed.

Product Overview
KTM 5 will support up to 16 CPU

cores!

Builder
Project Builder is the master configuration tool for setting

up both “invoice projects” and “standard projects.”

License Utility
KTM has its own license utility. It is

available from the “File” menu in Project Builder, and allows you to view your license values and what you are using for any project.
More about Project Builder follows…

2 - Product Overview
Project Planner provides an easy-to-use user

interface to create and manage document sets and configurations used for advanced classification and separation solutions.

Server
The Server may run as an application or as

a self-starting Windows service. Up to 16 services may be used. It classifies, separates and extracts. With the Pro Upgrade, you may also want to run separate instances of the Server: one for classification and separation and one for extraction.

Product Overview

Overview
The character-based Correction module allows for single-key correction of

characters read at a low confidence level. Thin client Correction is a new feature in KTM 5!

Validation
The Validation module allows you to review and correct

data that has been automatically extracted. You may also enter data manually.
Thin client Validation is also possible!

Overview
Thin client Verification is also possible!

2 - Product Overview
The KTM Knowledge Base Learning server

may run as an unattended application or a windows service. It accomplished sending documents “tagged” for online learning back to a pre-specified folder. It also instantly and dynamically updates Specific Knowledge Bases.

Base Administrator
The Knowledge Base Administrator is a standalone application

that is used to build training sets and create knowledge bases for an already-existing KTM project.

Classifier
The Image Classifier is an image classification and clustering

application useful for demonstrations.

Overview
The OCR Server allows you to create OCR output

(including Xdocuments) outside of Project Builder and without having to process batches through the Kofax Capture workflow.

Browser
The XDoc Browser provides direct access to the results

of automatic processing on a document-by-document basis. It may be used for testing and demonstration.

Viewer
The Statistics Viewer allows you to view summary statistics

of the results of automatic processing as you set up and modify your projects.

Capture Batch Workflow

Capture Integration
Easy-to-use Synchronization Tool integrates KTM projects with Kofax

Capture Document Classes and Form Types.