Category Archives for "Uncategorized"

How Intelligent Capture Utilizes Machine Learning for Template-free Training

There has recently been a lot of chatter in our industry about machine, and I want to share a few thoughts about how ImageSource’s ILINX Advanced Capture 8.5 platform tackles this topic. Our approach is to deploy a neural network-based document processing model that does not rely on templates. Our machine-learning platform supports custom-developed content classification projects with much faster turnaround than traditional rules-based models. The result is significantly faster time-to-production with more reliable and accurate results for our customer partners. ILINX machine learning offers:

  • Dynamic data location and extraction of information on complex documents
  • Image comparison to support check fraud applications
  • Feature extraction and classification to support medical imaging diagnostics

Our solutions leverage machine learning to create pre-built form classification algorithms in the lab, which provides a more-flexible and efficient way to develop new document processes.

For the sake of this article, I’d like to focus on the goals of almost every AP automation project:

  • Reduce paper handling and workload
  • Simplify processes down to one system for all invoices and other types of documents whether paper or digital
  • Gain visibility into where each invoice is in the process

For the initial document discovery, a technique called “clustering” can be used to automate the logical grouping of like documents. Clustering, in this example, refers to different categories of like invoices, checks, receipts and remittances. Documents can be organized automatically. Invoices from one vendor can be grouped together, such as receipts to travel documents. The result is a set of documents grouped by likeness that can then be further evaluated.

Next, each cluster, if part of a required document, can be given a document type (or class). The training set can then be imported into the machine learning ILINX solution designed to automatically identify key characteristics of each document type (often called “feature extraction”). This trains the neural network for each document type. When performance is not ideal for a specific class, the customer can add those misclassified or unclassified documents to the class sample set to “re-train” the neural network.

Data extraction is simplified by taking sample invoices that have been processed, along with the data required for each document. Together, these automatically train the software to locate the matching data and derive positional algorithms for each data field. The software uses the processed data for each page and locates the corresponding data on every document. The solution will do this for each sample and then automatically create algorithms based upon exact location, changes in placement across each example and relative position to other data, among other elements. The knowledge worker simply examines the results.

The technology used to configure the system also makes real-time adjustments. Complicated projects that typically would take weeks, if not months, are significantly reduced. Machine learning technology streamlines the manual processes used in production and helps reduce overall labor costs.
This effort can be applied to automate both paper-based and electronic document-based processes in a single workflow.

by Terry Sutherland, CEO, ImageSource

If you would like to learn more about how ILINX machine learning can automate your business please contact us at inforequest@imagesourceinc.com or

What was your “ah-ha” moment in communicating ECM?

Working in Content Management for over 12 years often times I have found it somewhat difficult to explain what we do and/or sell. Have you?
I have found that who your audience is often dictates how you explain it. To an IT group I have described ECM in terms of storage and retrieval of images in to database/repository with searching capability, ability to apply rules for authentication and accessibility, removing silos of information, ability to do workflow and BPM, and other things like Meta-Data, networks, through-put and HA/DR. Sometimes their eyes gloss over and other times they “understand.”
To some business folks when I’ m talking ECM I most usually reference things like accessibility of their documentation, being able to search on key fields and automatically route work/documents/content without the use of email or paper files (at its simplest form) and its all stored in a database otherwise known as a “repository.” Or, when describing workflow, using the old analogy of a restaurant. When you go in to the establishment a hostess seats you, then you get a menu, a waiter comes up and then you order, that order goes back to the kitchen and you get your meal prepared, then after you have dessert, you get a bill, pay and get a receipt then the bus boy comes and cleans everything up – that’s a workflow.
But what do you say to your mother or father, sister or brother and even children (aka the layman)? I’ve tried things like, “I sell software that lifts information off paper or documents and puts that data in a data base that allows people to find it. Then the people can see the documents on their computer necessary to do their job.” But I still get a ‘blank stare.’
Then one day, maybe three or four months ago, my dad was asking me for his usual P.C. help and he said, “my printer/scanner isn’t reading the words as well as it used to.” Of course, that got my attention! Could my dad know what O.C.R. is? After 12 years of me talking about IBM, FileNet, EMC/Documentum, Microsoft , Captiva, Kofax and AnyDoc and him saying “I still don’t get what you do.” NO WAY! How could my dad possibly know about O.C.R?
So I asked him, “Dad, you know what OCR is?” Guess what, he replied YES! “Its that software that I use when I want to take words off my documents that are PDF or Tiffs”. BAM! He knew! Finally after 12 years he “figured it out” partially what I did for a living. Putting this in context, my dad is an automotive guy, first sales and then executive, who had never a need to do any “computing” most of his professional career.
We have a lot of acronyms in our ECM vocabulary: OCR, ICR, OMR, BPM, OSR, ODAR, HIPI, TIFF, etc etc etc. (I can go on for a lifetime of our acronyms). But what do you say so that IT people get what ECM is? What do YOU say to a business user, who never ever ever thought of this stuff day to day? What do you tell your mom, dad, brother, sister, what you do every day? What have you said that brings blank stares? But, most importantly, what have you said to a customer and then you saw the “light bulb” go off? It appears O.C.R. is making it in to the mainstream vocabulary, if my dad is any example, because he knows his, “HP MFP does OCR.”

So I asked him, “Dad, you know what OCR is?”  Guess what, he replied YES!  “Its that software that I use when I want to take words off my documents that are PDF or Tiffs”.   BAM!  He knew!  Finally after 12 years he “figured it out” partially what I did for a living.  Putting this in context, my dad is an automotive guy, first sales and then executive, who had never a need to do any “computing” most of his professional career.

We have a lot of acronyms in our ECM vocabulary:  OCR, ICR, OMR, BPM, OSR, ODAR, HIPI, TIFF, etc etc etc.  (I can go on for a lifetime of our acronyms).   But what do you say so that IT people get what ECM is?  What do YOU say to a business user, who never ever ever thought of this stuff day to day?  What do you tell your mom, dad, brother, sister, what you do every day?   What have you said that brings blank stares?  But, most importantly, what have you said to a customer and then you saw the “light bulb” go off?  It appears O.C.R. is making it in to the mainstream vocabulary, if my dad is any example, because he knows his, “HP MFP does OCR.”

Document Scanning Best Practices

Document Scanning Best Practices Content Management Systems are one of the most useful resources companies have available to keep their managers, staff, and customers informed. Managing those files effectively is an ongoing challenge, but a well-planned, best practices implementation makes it significantly easier. Most Content Management Systems start with Scanning as the starting point in the lifecycle of any document. The decision of whether to go with a centralized or distributed scanning model must be carefully evaluated to see which may be a better fit for the organization. Many times a hybrid model of both remote and centralized is required and becoming more popular. When it is done designed and implemented correctly scanning ensures that the data stored in the document management repository is valid, readable, secure, accessible, and useful throughout the enterprise. Some important things to remember when deploying document a document scanning system: • Establish clear goals and objectives before you start or deploy a Document Scanning System. • Establish clear and concise business rules around your company’s requirements. • Consult a well established Systems Integrator with the knowledge and expertise to help you with defining “Best Practices for Document Scanning” and always check references. • Understand the nature of your documents, the quality of many documents may be poor, this in turn will require you to use Image Enhancement Technologies that will automatically clean up the document and improve its readability. These types of technologies are a must especially when utilizing OCR or any advanced form of capture. • Scanning and especially the Indexing of documents can be somewhat laborious, so anything to help automate these tasks such as Bar Coding, OCR, database lookups and electronic forms will make life a lot easier. • Use the KISS Principle in dealing with data taxonomy and avoid capturing too many fields, but make sure it’s enough to do valuable searches. Here at ImageSource we try to have 10 document types maximum and 8 data fields which allows for effective searches, retrieval and reporting. Lastly, don’t lose sight of your short and long term goals, do your homework and study your documents and see how they fit into your business lifecycle and corporate governance. Talk with people throughout you organization and get their input to better understand your documents are used. Finally, if you’re unsure get help, this is not an area where you can afford a mistake. Remember, it all starts with getting information into the system. Bob Garrido Senior Account Executive ImageSource

Putting Together an ECM Project Team Part 2

Part 2 – The Project Team

I discussed in the last blog on this general topic that the strong support of the intended Project by executive management is a critical factor for success – they need to support the projects sponsor, and smooth the path of challenges that sometimes occur when change is contemplated.  Vibrant and effective executive leadership is likely to be critical in solidifying the vision for the project.  The target of effort to achieve project acceptance and enthusiasm is cascading in that the focus of executive leadership is middle management, and then it effort fans out to focus on users and supervisors.

What Will be the Right Team

The right team of players, working together to hone the vision, is required to construct the concepts to be considered, refine the concepts, and to develop strategies to support the selected conceptual structure to fruition.  The people on the team are as integral to your project’s success as the solution, the project plan, the software tools, and infrastructure that is chosen.

Forming the right team is not easy – as not all leaders and users welcome new ideas and changes to the routine process.  But the executive support and the right team members are just as important for for standard ECM projects success as these factors are vital for business process management (BPM) and integrated implementations.

The primary key role types that are required on any ECM project team are listed below.  The exact position titles and numbers of team members recommended for participation will differ depending on an organization’s size and individuals’ skill levels. It is important that the eight classifications of people resources below are part of the Project team.

1. Executives: provide the supporting vision and enthusiasm for the solution objective

2. Line-of-Business (LOB) Managers: provide important project support and key higher level objectives

3. Business Analyst: provide discovery and analytical resources, reporting, perspective and ideas

4. Records/Compliance Manager: assure objectives and solutions match mandates and requirements

5. IT/IS Manager: supporting infrastructure, including business & IT challenges into the plan

6. WorkGroup Manager/Supervisor staff:  provide working knowledge of operations being addressed and realistic possibilities on what will work and where the challenges will be

7. End users: discovering what will and won’t work and where the challenges for acceptance are

8. Project Manager: the organization’s operational leader of the project and the coordinator with outside resources – ECM industry experts, software vendors, conversion resources, Training, etc.

From time to time this blog will continue with the subject of team challenges, some considerations to remember, use of supporting vendor resources, and some recommended methods for implementation.

Neil W. Lindsey, ECMM, CDIA+
Project Manager / Senior Business Analyst
ImageSource, Inc.

Enterprise Content Management Brings Efficiencies to Government

In a press release dated December 3, 2009, Governor Gregoire announces several significant reform measures to ensure Washington State delivers services more efficiently and effectively. One area of focus is information technology opportunities.

“Wide adoption of IT shared services allows agencies to concentrate on mission-critical needs while achieving savings through economies of scale and standardization,” said Tony Tortorice, state Chief Information Officer. “This is better for the agencies as they work with fewer resources to provide better service, and it’s better for Washingtonians who will receive better service at a lower cost.”

At Nexus 2009, the ECM solutions conference hosted by ImageSource, two government entities received ECM awards for improving services and creating efficiencies in their daily operations.

The recent ImageSource press release, dated November 10, 2009, highlights the two ECM solutions.

The Department of Financial Institutions (DFI) received the award for Best Use of Complementary Technology for a solution that automates the routing, monitoring and management of imaged checks and supplementary documents received by DFI.

“This solution assures that all checks and supplemental documents are imaged almost immediately, facilitating the rapid deposit of checks received by DFI.” said Ron Seymour, CIO, Washington State Department of Financial Institutions. “Being recognized at the Nexus conference highlights DFI’s commitment to develop technology that benefits the State of Washington,”

The City of Richland received the award for the Largest Return on Investment for an ECM solution that streamlines their agenda process by creating electronic workflows.

“Technology is rapidly changing how cities in Washington can gain efficiencies, improve processes, network and share information. Our objective is to make Richland a leader in information management,” said Jon Amundson, Assistant City Manager, City of

Richland. “We are honored to be recognized by this award which demonstrates the economical benefits this technology can bring to the City and its constituents.”

1 2 3