Today we’re pleased to announce a new release of IBM Watson Knowledge Studio, which enables experts and developers to teach Watson how to discover meaningful insights in unstructured text. To use Watson Knowledge Studio, you don’t need to have a PhD in artificial intelligence or even write any code. You train Watson, not program it. No matter what your field of expertise, you can train Watson using the intuitive cloud-based interface to identify entities and their relationships in unstructured data. You can build examples and collaborate to maintain consistency in how those entities and relationships are defined. makingrule-1 Based on your training, Watson Knowledge Studio then produces a machine learning model that you can use with Watson Discovery, Watson Explorer or AlchemyLanguage services to create applications that use and analyze content specific to a particular industry or domain.

What’s new?

With this release, the value of Watson Knowledge Studio has been expanded to meet more needs related to language, cost and data isolation. Here’s a list of what’s new:
  • Use the visual rule editor: Now you can create custom text extractors more quickly using business rules in Watson Knowledge Studio. You can use the visual rule editor to accelerate machine-learning model training, then deploy it to AlchemyLanguage, Watson Discovery Service or Watson Explorer for text extraction.
  • Work with more languages: Support is now available so you can work with documents in Italian, Brazilian Portuguese, French and Korean, adding to the languages that were already supported (English, German, Japanese, Arabic, French, Korean and Spanish).
  • Accelerate training: To speed up training, you can pre-annotate documents with Watson’s AlchemyLanguage entity extraction service, which finds mentions of entities in documents automatically. Pre-annotation helps you train custom models more quickly and with less effort than annotating manually.
  • Fine-tune results: You can iteratively fine-tune the results to meet your requirements. A new statistical tokenizer is introduced as the default. The previous dictionary-based tokenizer is available as an advanced option when you create a project.
Watson Knowledge Studio is also now available in two editions: Standard and Premium:
  • Watson Knowledge Studio Standard edition reduces the cost of entry by offering more flexible pricing options that can be tailored to your needs.
  • Watson Knowledge Studio Premium edition offers the same capabilities as the Standard edition with added data isolation. The premium edition also supports enterprise-wide deployments.
With Watson Knowledge Studio, you can create and deploy text annotators and provide them directly to AlchemyLanguage, Watson Discovery Service or Watson Explorer for building cognitive applications more quickly. A 30-day free trial of Watson Knowledge Studio Standard edition is available now.

3 comments on"New version of Watson Knowledge Studio meets needs related to language, cost and data isolation"

  1. sowmya suravarapu January 25, 2017

    I’m very glad to work on Watson Knowledge Studio.I have a few doubts like how to create multiple annotators in our project and some of the Enterprise level scenarios on watson knowledge studio.

  2. Swetha Gangadharan February 07, 2017

    We look forward to maximise the use of Watson in our project. For this purpose we have to educate waston with all project documents, User manuals etc and use the Watson Conversation API possibly. But this seems a manual process of creating entities, intents and then dialogues. Is there a way Watson can identify potential entities and intents on its own? Or some way of automating or fast tracking the process of identifying entities, intents and creation of dialogues.

    Can RPA (Robotic process automation) tools such as WinAuto or BluePrism be combined to use for this purpose ? Please suggest.


    Swetha Gangadharan

    • March 22, 2017

      Hi, Swetha. I would join the Slack Watson developer community and ask your question there.

Join The Discussion

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