May 10, 2016 | Written by: Mark Martin
Share this post:
We’ve recently released an update to Datacap which provide some interesting enhancements to our Cognitive Capture capabilities. With Datacap Insight Edition, we’re already using capabilities coming from Watson, IBM Analytics and IBM Research to look inside highly variable document types like correspondence, contracts or reports to find and extract important information and to make it available to users and applications for action. With this update, customers can now opt to use IBM Watson’s AlchemyAPI to go even deeper.
AlchemyAPI does many things. With Datacap Insight Edition we use it to find entity information inside documents. AlchemyAPI, is capable of identifying people, companies, organizations, cities, geographic features and other “typed” entities that are inside your documents. Entity extraction is based on sophisticated statistical algorithms and natural language processing technology. It is unique in the industry with its combination of multilingual support, context-sensitive entity disambiguation and comprehensive type support.
There’s more. AlchemyAPI sentiment analysis is capable of computing sentiment at many levels – the document, user-specified targets, entity-level, quotation-level sentiment, right down to the keyword level. These multiple modes of sentiment analysis provide for a variety of use cases and they can return information that leads to understanding the nuances and context of a document. These are just two examples of the broader analytic capabilities that AlchemyAPI provides to Insight Edition – capabilities that you can invoke from directly in line with your Datacap capture process.
So, why is this important? Remember the use case for Insight Edition – valuable, complex documents with variable formatting and content. These sorts of documents are difficult to automate with traditional imaging systems. How do you automate something that requires human understanding and logic to interpret? Well, now, you can use Datacap Insight Edition and take a big step toward that automation goal.
Imagine being able to automatically process incoming correspondence. Each letter is different – different structure, different formatting, and possibly even different languages. Today, this would require manual review to determine the content and to determine the next steps. But Datacap Insight Edition, drawing on AlchemyAPI, IBM BigInsights and other text analytics and language processing tools can automate and accelerate. It identifies entities in the letter and returns person, place and role information like “Joan,” “manager,” “Main Street” and “branch.” It can look for specific industry terms and return keyword-value pairs like “Account Number” and “12345.” It can determine actions and activity like “rude,” “ignored” and the sentiment score – positive or negative and what degree – and more. All this is done automatically and with no prior knowledge of the document. Net result: you know you have a complaint letter from a customer whose account number is 12345, who is in a negative frame of mind because she felt ignored by Joan, a manager who may have been rude, at your Main Street branch. And that is actionable information, information you can use to solve the problem, make a decision and respond correctly.
AlchemyAPI integration is the latest piece of capture innovation from the IBM Datacap team. It requires Datacap Insight Edition v9.0.1. Customers can download the Datacap v9.0.1 Feature Pack 1 from Fix Central here.