IBM Watson Assistant

Use IBM Watson Assistant to build a virtual assistant that understands natural language requests from users and responds with appropriate answers.

Important: On 30 September 2020, support ended for this application. For more information, see the withdrawal announcement. If you are interested in on-premises deployments of Watson Assistant, see the IBM Cloud Pak for Data documentation.

The Watson Assistant tool that is available in the IBM Cloud Private environment guides you through the steps of building training data that the service needs to build a machine learning model for you. The resulting custom model understands the types of requests that your customers typically make. It also provides graphical tools you use to build a dialog that the assistant can follow when it interacts with your customers. The dialog keys off the customer goals that you teach the service to recognize, and provides useful responses that are crafted by you.

The tool does not currently include integrations for deploying the finished assistant or metrics for analyzing conversations that your assistant is having with your users. Such deployment and metrics features are available from the public cloud version of the service only.