Chatbots: The hype, challenges and opportunities
summary
Argues chatbots excel at narrow, sequential tasks not general intelligence and shows exactly how to build and deploy one. Uses QnA Maker to extract QA pairs, the Microsoft Bot Framework and Azure to publish an endpoint, and a Zoopla API-backed demo to power a property search. Cites that ~85% of smartphone time sits in five apps and mobile messages jump from ~15T to ~30T, and that 80% of marketing leaders aim for bots by 2020. Builds an FAQ bot in five minutes and a working property search bot.
AI-written summary · about
chapters (15)
- 0:00Welcome
- 0:52What is a chatbot
- 1:23History of chatbots
- 8:26Customer service
- 9:58Opportunities for developers
- 10:59Chatbots for everything
- 13:36Personal finance
- 15:08How do you start
- 15:59What will your bot do
- 16:21Create your own bot
- 16:38Build your bot
- 27:08Design patterns
- 28:05Building a good bot
- 29:17How to disrupt the industry
- 30:26Second demo
related talks
56mBots - the next UI revolution
47mBuilding Intelligent Enterprise-Grade Help Desk Bots using Conversational AI
36mChatbots Past, Present and Future
42mLet’s Chat-Bots using Azure AI
29mAvoiding the Chatbot Trap Designing AI Interfaces That Unlock Real Business Value
55mEnterprise Chatbots (using messaging and AI to build novel user interfaces for work)
35mDesigning Experiences for Chatbots
59mHow to create No Code Chatbot with Azure QnAMaker