Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chatbots - Coggle Diagram
Chatbots
Why use it?
-
-
With the use of chatbots, manpower can be re-located to areas of work which cannot be automated
-
-
Entities
In Dialogflow, there are system entities- e.g. dates, times numbers, email addresses, etc. You can also create your own custom entities. For example, you could define a vegetable entity that can match the type of vegetables available for purchase with a grocery store agent.
parameters and entity types are structured data that can help us make better sense of what the user is trying to find out.
-
Agents
A visual agent with the mission to handle conversations with your customers. Once a user send a text message to the Dialogflow agent, Google uses AI to translate the message into structured data with the aid of NLP.
Dialogflow agent can handle human language. Dialogflow agent needs to be built and trained to reply to conversations.
Intents
Follow up intents
This is because real life conversations are more than a simple question-and-answer. After our chatbot provides an initial response to a customer's question, we can expect the customer to follow-up from that response
An intent categorises the customer's intention for a conversation turn. You can define multiple intents for a single agent, just like how a call centre agent can decide if you need basic troubleshooting, make an appointment, or redirect your phone calls based on what you say
How does it work?
A dialogflow conversation begins from the (1) end-user expression, have(2) its intent matched by the agent with the use of training phrases and entities, before(3) response is being provided back to the user.
Google dialogflow
It uses National Language Processing (NLP), which is a branch under AI which we can train computers to understand and process the natural human language.
Human conversations are fluid and can sometimes be unpredictable; hence it is important to continuosly train our chatbot.
While our chatbot has been trained to match intents and entities, we hadn't teach it how to respond to a simple "Thank you"- and phew! We don't have to do it manually.