Building Bots with Watson Conversation
Download as PPTX, PDF10 likes2,615 views
This document discusses bots and chatbots. It defines bots as software applications that run automated tasks over the Internet, typically performing simple and repetitive tasks. Chatbots are described as computer programs that conduct conversations via text or audio. The document then provides examples of bots and chatbots, and discusses some of the key concepts for building a bot using Watson Conversation, including intents, entities, context objects, and dialog structure.
Building Bots with Watson Conversation
- 1. © 2015 IBM Corporation Stefania Kaczmarczyk Developer Advocate No Code Bot Building with Watson Conversation
- 2. 2Page© 2015 IBM Corporation “An Internet bot, also known as web robot, WWW robot or simply bot, is a software application that runs automated tasks (scripts) over the Internet. Typically, bots perform tasks that are both simple and structurally repetitive, at a much higher rate than would be possible for a human alone.” - Wikipedia Bots “A chatterbot (also known as a talkbot, chatbot, Bot, chatterbox, Artificial Conversational Entity) is a computer program which conducts a conversation via auditory or textual methods. Some chatterbots use sophisticated natural language processing systems, but many simpler systems scan for keywords within the input, then pull a reply with the most matching keywords, or the most similar wording pattern, from a database.” - Wikipedia Chatbots
- 3. 3Page© 2015 IBM Corporation Lyft it from Facebook
- 4. 4Page© 2015 IBM Corporation An AI teaching assistant for an artificial intelligence class at Georgia Tech “Jill Watson”
- 5. 5Page© 2015 IBM Corporation But first! Get a service deployed. Hint: Go to bluemix.net
- 6. 6Page© 2015 IBM Corporation Now start bashing out some intents.
- 7. © 2015 IBM Corporation Just kidding. No code.
- 8. 8Page© 2015 IBM Corporation You Know Nothing of Language
- 9. 9Page© 2015 IBM Corporation Feedback and Repetition – Don’t ask the speaker to repeat themselves too often – Confirm details: “Your order was made on September 1, 2016 right?” – Get to the point and take the short route when appropriate – Remember details about them
- 10. 10Page© 2015 IBM Corporation Intents – What’s the weather? – I need new glasses – Where is the closest pizza place? What a user wants to do even if they don’t know that they want to do it. ere
- 11. 11Page© 2015 IBM Corporation Entities – Pizza toppings – Car makes – Articles of clothing Nouns and their synonyms. Adding variation that the bot can narrow down to a specific choice and store.
- 12. 12Page© 2015 IBM Corporation Dialog – conversation_start – Anything else
- 13. 13Page© 2015 IBM Corporation Testing the Bot
- 14. 14Page© 2015 IBM Corporation Brainstorm a Bot – Scope – avoid scope creep – Start a conversation with someone else and see how they ask questions – Pick an industry or a tough task like calling tech support – Imagine a life without other humans
- 15. 15Page© 2015 IBM Corporation Getting Advanced: Context Objects – Store user selected variables – Store details about the user – Feed information to your app – Turn the user into their own SQL search builder
- 16. 16Page© 2015 IBM Corporation Recall – Directly recall or store the user’s text: <? Input.text ?> – Echo or store an entity: @coffee – Recall a context variable: <? $username ?>
- 17. 17Page© 2015 IBM Corporation Loop Back – Reroute the user to another conversation path – Avoid repeating logic – Reprompt
- 18. 18Page© 2015 IBM Corporation Next Steps – Connect the conversation service to an app – Store responses in a database and look for edge cases – Connect it to an actual robot For a basic chat dialog with Node.js fork the code at: http://bit.ly/2djEkfa
- 19. 19Page© 2015 IBM Corporation Credentials!