Machine Learning for Developers / Collection

Smart Chatbots

Watson Conversation is smart. Watson Conversation with data is smarter.

Mark Watson
Welcome to my data lab.
More by Mark Watson

If you have a website for your business, what are the most popular inquiries it generates?

  • When are you open?
  • Can I make a reservation?
  • Do you deliver to my address?

Many companies are now creating chatbots that can answer such questions. The human has an interactive conversation with the bot on the company website, and information is passed to the customer in a very natural way:

Screen shot of a hotel chatbot window close up, in three panels.

Services like Watson Conversation let you program chatbots that respond in a human-like manner. The deeper you get with your natural-language-processing and deep-learning techniques, the richer the interactions, and the more insightful they are.

Getting the most out of your chatbot interactions also means persisting data for analytics. That means understanding your database options, and we’re here to help. There are lots of ways to do it. For instance, this example hotel chatbot uses serverless functions on Apache OpenWhisk, dialog trees in Watson Conversation, and the IBM Cloudant database service:

Screen shot of a hotel chatbot application, in the context of a mock website.

From graph data models to enriching chats with API metadata from previous interactions, the projects in this collection will help you get started building smarter chatbots.

↓ View projects in this collection

Projects