Alphabet Soup

Until a few years ago computers were lousy at understanding concepts in text or dealing with complex decision criteria.

Opportunities through advances in NLP

Until a few years ago computers were lousy at understanding concepts in text or dealing with complex  decision criteria.  The capabilities of Natural Language Processing (NLP) have, however, evolved rapidly over the last 10 years providing tangible opportunities to deliver real business value with NLP solutions.

Roughly stated, we have moved from automated counting and matching of words  to a deeper understanding of words, relations and concepts in text. Until recently, for example, computers were only able to find and count the words (e.g., ‘money’) as well as closely related words but didn’t have a clue about broader concepts (e.g., a means of exchange, multiple currencies, savings, rates etc.). For us humans it is simple. We easily grasp concepts and combine these with other concepts and context. Computers need our help to accomplish the same feat, and thankfully the tools to provide that aid are slowly becoming available.

The precise technical details are beyond the scope of this article, but suffice it to say there is a lot of technology to enable better NLP but  fundamental knowledge on how to process language and implement interaction models is far more invaluable. More importantly, what can you actually do with advanced NLP solutions? Cases where people with  domain knowledge make interpretations and decisions to drive a process to an outcome are in many ways a natural fit for advanced NLP solutions.

Such as:

  • (Pre)processing Letters of Credit in a bank
  • Due Diligence on agreements, contracts and portfolio’s
  • Screening and summarizing legal documents for a case
  • Governance of procurement contracts
  • Assessing large real estate portfolio’s and (syndicated) loan agreements
  • A versatile tax-bot guiding you through a vast legal database

These are all examples of processes that require a certain understanding of the domain or  a frame of reference. Also, they often involve a vast amount of text with many concepts, relations and reference checks. Since human expertise is difficult to scale, using NLP can prove invaluable in efforts to automate large portions of these processes and, thus, truly scale knowledge work. 

We deliver solutions that work faster, more consistently, and are more compliant, and cost-effictive. Our solutions operate in close collaboration with real human experts in order to deliver the best of both worlds. We help define the business case, the solution and take care of the delivery. For more information about our solutions  see: ( ) Curious on how this will work for you? Plan a meeting or a demo.