IBM Watson Knowledge Studio – Teach Watson about your domain

Share this post:

Author: Tanmay Sinha and Andrew Boyd

Unstructured Data – An Untapped Opportunity

Every day we produce 2.5 quintillion bytes of new data – the majority of which is unstructured. While this data holds many possibilities, it is vastly underutilized because of the complexity and challenges in using this data. In fact, typical organizations only leverage 8% of this data.

Screen Shot 2016-06-24 at 2.44.45 PM

Most critical of this unstructured data is in natural language text e.g. books, journals, health records, e-mails, blogs, tweets etc. Not only are we producing a vast amount of this natural language text daily, but the rate at which we are producing the text is growing – so, there is great opportunity for savvy organizations to leverage this text to make game-changing decisions and provide business value.

How can your organization tap in to all that natural language text for business insights?

Introducing IBM Watson Knowledge Studio

IBM Watson Knowledge Studio is a cloud-based application that enables developers and domain experts to collaborate on the creation of custom annotator components that can be used to identify mentions and relations in unstructured text.

The effort behind IBM Watson Knowledge Studio began almost two years ago when a joint team from IBM Research and the IBM Watson Group did a study to understand existing natural language processing (NLP) technologies and best practices. The goal of the study was to recommend a common direction for tooling, repositories, data formats and adapters to enable interoperability across IBM’s cognitive applications and reduce the time needed to customize them.

Knowledge Studio has come a long way in the past couple of years, but our founding vision to establish interoperability and simplify domain adaptation to create cognitive applications has remained constant.

The traditional approach to adapt, or train, a cognitive application requires the subject matter expert to know NLP concepts, such as understanding syntax for creating rules. But not anymore. With Knowledge Studio, we are reinventing how you can teach Watson and create cognitive applications specifically for your domain.

Understanding Natural Language Text

Systems like IBM Watson provide Natural Language Processing power to “understand” unstructured text. The problem, however, is that most organizations are looking to mine unstructured text for specific information and context that are unique to their industry or business needs. As a result, organizations must customize the NLP model in order to realize the full value/benefit of mining the unstructured text.

For instance, imagine you are an auto manufacturer; you’re looking to proactively identify safety defects using traffic incident reports. To achieve this, you would need to create a NLP model that – through context – understands manufacturer, make, model, type of incident and date of incident.

Screen Shot 2016-06-27 at 9.24.08 AM

While the example is straight-forward, the process of customizing the NLP model can be a cumbersome process that requires a rate set of specific skills. The good news is that it doesn’t have to be.

IBM Watson Knowledge Studio dramatically lowers the bar for customization of natural language processing models – what’s more, through innovations in underlying technology, the process is much faster than traditional approaches.

Watson Knowledge Studio is:

  • Intuitive: Use a guided experience to teach Watson nuances of natural language without writing a single line of code
  • Collaborative: SMEs work together to infuse domain knowledge in cognitive applications
  • Cost effective: Create and deploy domain knowledge infused annotators faster than ever before using an integrated development environment

Unlike tools from the past, Knowledge Studio is built from the ground up for subject matter expert and developers to collaborate. It leverages decades of IBM experience in NLP and applies the IBM Design Thinking approach to distill the training process down to the most user-friendly experience possible.

The various components of Knowledge Studio come together to address all aspects of domain adaptation for unstructured text – starting from defining a type system to deployment of a fully trained model directly to offerings IBM Watson Explorer and AlchemyLanguage.

IBM Watson Knowledge Studio and AlchemyLanguage – Custom Real-time NLP on Cloud

Since AlchemyAPI joined IBM Watson, we’ve continued to enhance the capabilities of our platform.

As part of that effort, we are excited to announce that you can now “train” a custom model about your industry or specific knowledge domain using IBM Watson Knowledge Studio. Here is how you can to get started creating and using your custom model:

  1. Purchase Watson Knowledge Studio
  2. Get your Alchemy API key from Bluemix – create your key using the Advanced plan
    Train and publish a model
  3. Use your model through the API

IBM Watson Knowledge Studio and Watson Explorer – All enterprise text analytics needs addressed in an integrated solution

Watson Explorer is a technology platform that accesses and analyzes structured and unstructured content. Explorer presents data, analytics and cognitive insights in a single view and gives customers the information they’re looking for while uncovering trends, patterns and relationships.

With the release of Watson Explorer 11.0.1, customers can utilize machine learning models developed using Watson Knowledge Studio to annotate and enrich content on premise. Watson Explorer is the first and only IBM offering to support a hybrid approach that allows both custom rules and custom ML annotators for annotating unstructured text.

Try IBM Watson Knowledge Studio Today

IBM Watson Knowledge Studio dramatically lowers the time and expertise needed to train a Watson solution for a particular domain or area of expertise – what was once measured in years is now measured in weeks. Most importantly though, you are teaching Watson, not programming it.

IBM Watson Knowledge Studio represents an innovative new approach to training natural language processing models and opens up new possibilities for cognitive computing across industries.

Questions? We are always here to help. Reach out to us via our support forums.

IBM Watson Knowledge Studio is available on IBM Marketplace.

Add Comment

Leave a Reply

Your email address will not be published.Required fields are marked *

Aaron Tarin

It’s been well over a decade since I first beheld natural language parsing as an undergraduate studying the emerging (comparatively primitive) machine translation algorithms. They said the ontology frameworks of natural languages were simply too dynamic, vast, disparate and chaotic. We thought they were perhaps simply too, human…to ever be understood by machines. Many laughed backed then, as they occasionally still do now. But the laughter has largely dissipated to gasps, as even lawyers come to the realization of what is now upon us.


Rishi Astra

Is there an article explaining more integration between WEX and WKS?


Munmun Nath

How can IBMers get access to Knowledge Studio for our internal projects?


    Lisa Kay Davis

    Thanks, Munmun. Look out for an email shortly with an update.


Alexis Rodriguez

Can a customer / an IBM facing a cognitive project use WKS as tool to train WEA?


Alexis Rodriguez

Can a customer or an IBM use WKS to build a specific knowledge corpus underneath a specific WEA implementation?


    Tanmay Sinha

    Alexis – At this time, WKS is not integrated with WEA.


Shamail Saeed

Easy to use interface, one can watch YouTube video and tutorial on IBM site and can immediately start using it in team mode or individual mode.

It would be better if they provide some pre-trained models so that we don’t have to train with a lot of like 200 docs.

More Developers Stories
December 4, 2017

Freedom and flexibility with Speech-to-Text

When it comes to speech-to-text solutions, an out-of-the-box service isn't enough. IBM Watson's Speech-to-Text service helps provide the tooling and functionality to train Watson to learn your business.

Continue reading

December 1, 2017

The future of AI Revenue: Top 10 use cases for AI in the next decade

6 AI segments will account for a significant change in revenue for AI enterprise software. These functional areas are applicable to many use cases, industries, and generate benefits for both businesses and individuals. Here are the top ten use cases which will reap financial rewards for AI technology product and service companies, and a broad spectrum of benefits for everyone else.

Continue reading

November 20, 2017

How fund managers can apply AI to turn data into insights, reduce bias in decisions and generate alpha

In this age of rampant data growth, the only way to reliably beat the market on a risk-adjusted basis is to mine unstructured data faster and more accurately than competitors. Companies that combine AI, and machine learning with speed, accuracy, nuance and contextual awareness will change the game of managing and growing investments.

Continue reading