Deconstructing Chatbots: An Overview
I have been in the chatbot world for a long time. When I joined Google Cloud team, one of my interests was to focus on Dialogflow and soon I started to get pings from developers and business leaders who wanted help getting started with Dialogflow and chatbots. So, I decided to start this focussed chatbot series called “Deconstructing Chatbots” where I plan on sharing all my learning with you.
For us humans, Conversation is natural, it is part of our everyday life. We fundamentally understand it, and all the nuances around it, because honestly it’s part of who we are. This is why trying to teach a machine to have a conversation is so difficult.
How you interact with machines right now seems pretty simple. You just ask something, and the machine responds. But it turns out that conversation is a really hard thing to get right for a machine, since people ask for information in different ways.
For example, to do something as simple as get the weather, you could say “What’s the forecast like today” or “What’s the weather right now” or “what’s the temperature like in san fran tomorrow
If you were to code this, traditionally — you’d most likely need a whole matrix of conditionals to figure out all the edge cases for all the ways the user can ask for this single thing. But as you can imagine, that’s not really maintainable. This is where NLU (or, natural language understanding) helps.
NLU is a technology that helps translate human language into computer language, and visa versa. It’s very similar to Natural Language Processing, but it goes a step further to understand conversations that it hasn’t been trained to understand, like errors, spelling mistakes, accents, sentiments, etc. This makes NLU a great fit for chatbots. Additionally, NLU works with both voice, and text.
A conversational experience may not be on top of your list when you’re thinking about customer support, but it’s actually a critical element to meet customer expectations in a very personalized manner. With the changing times, users of any product or service are looking to get support via voice or chat, and..Google’s Dialogflow makes this possible.
Dialogflow lets you build conversational interfaces on top of your products and services by providing a powerful natural language understanding (NLU) engine to process and understand what your users are looking for. Which means when you use Dialogflow you get all the value of an NLU engine for your chatbots, without having to lift a finger.
In this article we covered conversation experiences, why they are important, NLU, and how Dialogflow lets you get up and running really quickly. In the next article, we are going to describe some critical terms in the chatbot world which will lay a solid foundation for us to start building an actual chatbot. So, keep following deconstructing chatbots.
