IBM Applies Cognitive Computing to Help Businesses Extract Critical Information and Insight from Documents

IBM Datacap Insight Edition applies imaging, analytics, and machine learning techniques to understand content and context

IBM INSIGHT 2015, LAS VEGAS - 26 Oct 2015: IBM (NYSE: IBM) today announced a new data capture solution – embedded with cognitive computing capabilities – to help businesses capture and extract more information and insight from enterprise documents.

IBM Datacap Insight Edition uses cognitive computing capabilities to expedite processing of unstructured, highly variable documents. (Credit: IBM)

Businesses today must contend with an unprecedented volume and variety of data, with unstructured information accounting for about 90 percent of enterprise information[1]. This comes in many forms, from documents to images, and rich media. And many day-to-day operations are driven by business transactions that produce paper and digital documents that companies must manage with capture and imaging solutions. As businesses complete more complex transactions like processing loan applications or cross-border shipments, human intervention is needed to review, classify and make sense of the document’s content.

By applying a combination of advanced imaging, natural language processing, and machine learning technologies, IBM Datacap Insight Edition can automatically classify and understand the document – including format and structure, as well words and numeric information – from any document type to help businesses quickly and accurately determine the appropriate action to take.

This is particularly important in:

  • Banking and Finance - where different business lines such as Retail/Commercial banking, Lending, and Investment banking require similar documents for differing contextual reasons
  • Healthcare - where doctors and hospitals are transferring hand written notes and images into electronic health records for analysis or filing.
  • Insurance - where roughly half of all documents received are titled “correspondence” and need to be analyzed for content and classified appropriately.

"In the insurance industry, it is important to accurately capture and understand unstructured data. This is especially true when it comes to claims because they often include a high percentage of unstructured data in the form of correspondence. A cognitive capture solution would be extremely useful in helping us to automate the processing of these documents," said Paul Deffinger, Director ECM Applications at Great American Insurance.

By analyzing the content, IBM Datacap Insight Edition can identify which content to apply analytics to at the point of capture, while continually learning about new document types for future use and overall faster processing. It applies business rules to help organizations determine what needs to be done – whether a document should be passed to a line of business system or should initiate a case and workflow for additional actions.

Integrating cognitive capabilities within document capture helps eliminate the need for time-consuming and costly manual intervention while maintaining the accuracy and speed required by businesses,” said Beth Smith, GM, IBM Analytics Platforms. “And even more importantly, cognitive capture allows a business to learn as it processes more transactions to continuously improve speed and accuracy, while making evidence-based decisions around critical day-to-day tasks and interactions.”

For more information on IBM Analytics, please visit

[1] Big Data, Bad Data, Good Data: The Link Between Information Governance and Big Data Outcomes,” Melissa Webster, IDC, January 2015.

Related resources


IBM Datacap Insight Edition uses cognitive computing capabilities to expedite processing of unstructured, highly variable documents. (Credit: IBM)

Related XML feeds
Topics XML feeds
IBM Analytics
News about IBM solutions that turn information into actionable insights.