IBM’s Jean Lange takes you through the process of training Watson, including building a customized machine-learning model and using services to teach the system to recognize and extract domain-specific content and insights.

In this video:

  • Jean Lange, Developer, IBM Watson Discovery Service

IBM Watson can extract helpful insights about your data out of the box. Like a knowledgeable friend, it “reads” through data to show you its themes and important points, but it doesn’t know everything. You have to teach it about your domain to get to even more helpful and relevant information.

In this webinar, developer Jean Lange walks you through the process of teaching the system something it doesn’t already know. She will explain how you can:

  • Create a (tiny) custom model using Watson Knowledge Studio
  • Deploy that model to the Natural Language Understanding service so Watson can recognize domain-specific content in unstructured text
  • Deploy the same model in Watson Discovery Service to show you how to extract domain-specific insights that are not possible right out of the box

The Knowledge Studio makes it easier for you to create a machine-learning model that understands the linguistic nuances, meaning, and relationships specific to your industry or to create a rules-based model that finds entities in documents based on rules that you define.

The Natural Language Understanding service allows you to add the ability to your application to let you perform natural language processing that can enable advanced text analysis capabilities.

The Discovery service allows you to add a cognitive search and content analytics engine to apps to identify patterns, trends, and actionable insights that lead to better decision-making. You can unify structured and unstructured data with pre-enriched content and use a simplified query language to eliminate the need for manual filtering of results.

Resources for you

Building with Watson

The one-stop shop to jumpstart your cognitive application development.

Join The Discussion

Your email address will not be published.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