When you provide a few example questions (intents) and clarifying terms (entities), Watson Assistant can build a machine learning model to understand similar natural language requests from your users.
Anyone can use the visual dialog builder to build responses to user questions and clarification flows into a skill — no developer experience required.
When a user changes the topic in a chat, the system will automatically leave the main subject, address the new topic, and then seamlessly return to the original subject.
When the user poses an ambiguous question or request that could have multiple responses, the system will automatically ask for further clarification, rather than jump to conclusions too quickly.
We've made it easier to identify new topics from pre-existing chat logs. With Intent Recommendations, simply import chat logs and Watson Assistant will recommend new intents.