Building Chatbots with API.AI - GDG Mumbai Meetup

Building Chatbots with API.AI - GDG Mumbai Meetup

Transcript

1. Building CHATBOTS with @MaunashJani
2. What is a CHATBOT? @MaunashJani
3. CHATBOT  A service - made up of rules and AI  User interacts with a chatbot via a chat Interface.  Platforms supporting Chatbots- Facebook Messenger, WeChat, Slack, Telegram, Text Messages, etc.  Aka - talkbot, chatterbot, Bot, chatterbox, Artificial Conversational Entity  Types – Command Based and Learning Based @MaunashJani
4. Examples  Weather bot  Order food delivery  News bot  Education  Buy movie tickets  E-commerce bot  Doctor’s appointments  Customer Support  And many more… @MaunashJani
5. HOW? AI, ML & NLP @MaunashJani
6. • Conversational User Experience Platform. • Once you build your agent, you can integrate it with SDKs or use our one-click integration modules. • When your product is launched, it constantly improves with machine learning and can be updated in real-time based on user interactions. • API.AI => NLP + Machine Learning + AI + Conversation building features @MaunashJani [Source: https://api.ai]
7. Get STARTED  Define Conversation  Create API.AI Agent and Intents  Build and deploy Webhook fulfilment  Define actions  Submit for Approval [Source: https://developers.google.com/actions/apiai/ and https://docs.api.ai] @MaunashJani
8. API.AI Basics  Agents  Intents  Entities  Actions  Contexts  Webhook @MaunashJani [Source: https://docs.api.ai]
9. API.AI Basics  Agent – NLU modules for applications to transform natural user language into actionable data. Agents can be designed to manage a conversation flow in a specific way.  Intents – Developer-defined components of agents. An intent represents a mapping between what a user says and what action should be taken by your software. Intent interfaces have the following sections:  User says, Action, Response, Contexts  Webhook - Webhook integration allows you to pass information from a matched intent into a web service and get a result from it. @MaunashJani [Source: https://docs.api.ai]
10. Entities  Entities represent concepts and serve as a powerful tool for extracting parameter values from natural language inputs.  There are 3 types of entities:  system (defined by API.AI),  developer (defined by a developer),  user (built for each individual end-user in every request) entities. @MaunashJani [Source: https://docs.api.ai]
11. CONTEXTS  Contexts are strings that represent the current context of a user’s request.  This is helpful for differentiating phrases which may have different meanings depending on the user’s preferences or geographic location, the current page in an app, or the topic of conversation. @MaunashJani [Source: https://docs.api.ai]
12. DEMO [CUSTOMER SUPPORT – SAMPLE DATA] @MaunashJani
13. 1. CREATE AN API.AI AGENT @MaunashJani
14. Ready INTENTS @MaunashJani
15. CREATE INTENT @MaunashJani
16. @MaunashJani
17. Next  Agent Training Completed @MaunashJani
18. Create Entities @MaunashJani
19. Integrations @MaunashJani
20. @MaunashJani
21. @MaunashJani
22. FB Messenger @MaunashJani
23. Summary - Create a chatbot with API.ai 1. Create an API.AI agent 2. Create Intent 3. Create Entities 4. Link Actions and Generate Response 5. Integrate with Messenger 6. Create a Facebook Page and a Facebook App 7. Get a Page Access Token 8. Subscribe the App to the Page 9. Configure with API.AI settings 10. Setup Webhook 11. Host on Cloud – Heroku/other host using the Verfication Token 12. Send/Receive Messages @MaunashJani
24. RESOURCES  Difference between AI, ML & NLP - https://goo.gl/glZGHg  API.AI - https://api.ai/  API.AI Docs - https://docs.api.ai/  Actions on Google and API.AI - https://developers.google.com/actions/apiai/  Getting Started with Google Actions with API.AI https://goo.gl/Y9YLyf  API.AI - YouTube Channel @MaunashJani
25. THANK YOU Maunash Jani Developer Evangelist @Genius_Lynx Trainer @Diplomads MaunashJani maunash@geniuslynx.com