Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chatbots - Coggle Diagram
Chatbots
Dialogflow
Natural Language Processing (NLP)
build, test, launch and improve
Intent
represent the various ways users can express their intentions or make requests.
Define the particular responses to the inputs
end-user expression - response
Default intents
Default welcome intent
First intent you see when starting up your agent
Default Fallback intent
Default response you see when agent cannot comprehend input
Follow up intent
To get more specific into a certain topic
Used to handle subsequent responses after the first intent
Can only trigger after first going through the original intent
you can have multiple follow up intents stacked
Entity
extracted data
represent specific details
Agent takes particular note of these keywords
Agent
can handle human language
needs to be built and trained to reply to conversations.
You can create multiple agents
core component of Dialogflow
Training Phrases
pattern and expression recognition
helps to improve understanding of the agent
Integration
Multiples online messaging platforms available to be integrated
Required if you want to add your bot
the process of connecting your chatbot with an external platform
Telegram
Requires a token key
Improving
Small talk
Validation
Training
Why
allow organisations to interact with customers 24/7
manpower can be re-allocated to areas of work which cannot be automated
help to retrieve and provide information to customers quickly
used for data analysis to boost sales subsequently