Frequently asked questions
Get answers to the most commonly asked questions about this product
What is Watson Knowledge Studio?
IBM Watson Knowledge Studio (WKS) is the ultimate tool for customizing your natural language processing (NLP) needs. It is a cloud-based application that enables developers and domain experts to teach Watson the language of their industry. Users train Watson by example, annotating sample documents with the information relevant to their domain. WKS is then able to train a machine learning model that can analyze new, unseen documents, thus scaling the knowledge of the domain experts.
How do I train Watson? What does that mean?
Watson Knowledge Studio uses machine learning to create domain-specific custom models, based on examples provided by the user. Users will select sample documents for their domain and define entity types and relationships among them. For example, a user might create entity types for "CarModel" and "Year," and a "MadeIn" relationship to connect them. To train Watson with WKS, you just have to annotate documents for it. With enough examples, your custom model will be able to analyze previously unseen text in the same way.
What is a type system?
A type system defines things that are interesting in your domain content that you want to label with an annotation. The type system controls how content can be annotated by defining the types of entities that can be labeled and how relationships among different entities can be labeled.
What is a rule-based annotator?
A rule-based annotator can recognize repetitive patterns in your documents, like license plates, phone numbers, dates, times, or phrases like "35-year-old driver." Use rules to capture patterns that occur in documents and convey information about underlying entity types. Consider using a free web-based regular expression utility to help you determine the right expression to use to capture the pattern of interest.
What is preannotation, and how do I leverage it?
Preannotation relies on user-defined dictionary, rule-based, or machine learning-based annotator components that you can run to find and annotate mentions automatically. For example, if your documents contain reviews of food products, you could upload a CSV file listing hundreds of foods, and have them automatically highlighted before your human annotators begin their work.
What languages are supported in Watson Knowledge Studio?
The Watson Knowledge Studio user interfaces, messages, and documentation are available in English only. However, you can train models in English (the default language), Arabic, Brazilian Portuguese, Chinese (Simplified), Dutch, French, German, Italian, Japanese, Korean, and Spanish.
What options are available for utilizing my Watson Knowledge Studio custom model? How do I manage my models?
Watson Knowledge Studio models can be deployed directly into Watson Discovery, as well as Watson Natural Language Understanding. They can also be exported into Watson Explorer, but this option is not available for Lite edition users. Watson Knowledge Studio includes a model management console, where users can easily deploy and undeploy their models.
How are Watson Natural Language Understanding, Watson Discovery, and Watson Knowledge Studio related?
All three products can be used independently of one another. Natural Language Understanding is a best-of-breed text analytics service that can be integrated into an existing data pipeline. Watson Discovery incorporates Natural Language Understanding, but also offers storage of data with multiple ingestion options and provides the opportunity of powerful queries to extract insights, serving as a powerful cloud native insight engine. Watson Knowledge Studio can be used with both Discovery and Natural Language Understanding to customize text extraction of entities and relations using machine learning models.
What security certifications does Watson Knowledge Studio have?
All Watson Services are now officially ISO27001, ISO27017 and ISO27018 certified, and GDPR compliant. PII can now be officially accepted in all deployment models (Standard, Premium & Dedicated) due to the ISO27108 certification. The only regulated PII Watson does not officially support is PHI (HIPAA) and PCI as these require their own certifications. Please read this page for further IBM Cloud security information.
Is WKS on Marketplace being replaced by WKS IBM Cloud? What will happen with my WKS projects on Marketplace?
Yes. WKS projects will be available for a limited time, depending on the edition of WKS being used. Please refer to our migration page for information about how to move your WKS projects to the IBM Cloud. There, you can find multiple resources (documentation, video, and blog post) that guide you through the migration process. Depending on the edition the customer is using, their instances will be shut down sometime in 2018. Please see our documentation for specific details.