Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chatbot - The 24/7 Agent - Coggle Diagram
Chatbot - The 24/7 Agent
chatbots are generally 70-100% accurate, which means they are pretty accurate
chatbots are consistent in their response - either they are consistently right or consistently wrong
-
-
-
2) with the use of chatbots, manpower can be re-allocated to areas of work which cannot be automated
-
-
5) chatbots can respond when customers ask questions about your product, even when you're sleeping
6) bots can retrieve information and respond quickly (within milliseconds so customers can be served in a shorter time)
-
-
-
-
Dialogflow uses Natural Language Processing (NLP), which is a branch under artificial intelligence. In particular, NLP is how we can train computers to understand and process the natural human language
-
1) Dialogflow allows you to build, test, launch and improve your chatbot
-
3) Dialogflow requires minimal coding and is suitable for anyone who can effectively use a computer and web broswer
-
5) Dialogflow is powered by Google's machine learning and has the cognitive ability to understand and reply in the natural language
-
-
once a user sends a text message to the Dialogflow agent, Google uses artificial intelligence to translate the message into structured data with the aid of NLP
customers reach out to chatbot agent to acquire certain information. hence, the Dialogflow agent needs to be built to handle the conversation
a Dialogflow agent needs to be trained to handle conversational scenarios. training does not need to be overly explicit
-
if the agent accurately identifies the user's intent, it is almost a success
an intent categorises the customer's intention for a conversation turn. you can define multiple intents for a single agent
what a customer types to the chatbot are referred to as an "end-user expression". Dialogflow matches the end-user expression to the best intent in your agent. matching an intent is also known as intent classification
the agent has to identify the intent, and use entities to piece together what the user is trying to say
-
when an intent is matched, Dialogflow can extract specific information from the end-user expression. the extracted data is called a parameter or entity type
parameters and entity types are structured data that can help us make better sense of what the user is trying to find out
in Dialogflow, there are system entities that are already present but you can also create your own custom entities
-
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
-
-
-
-
-
-
-
-
the token from botfather from telegram should only be provided to a trusted someone who needs to work on the chatbot integration
we can improve our agent through small talk, validation and training
you can teach our agent how to reply to phrases like "thank you" and "you look great today" or even customise small talk
Dialogflow provides a validation feature to help us create high-quality agents. agent validation results are available automatically whenever agent training is complete. these enable us to know if there are any issues with our agents
you can see all the conversations made with your chatbot, and manually train your chatbot better in the Training section (note: All conversations can also be accessed from the History button)
for each end-user expression, if the agent correctly matches the intent, you can click on the tick icon. this allows the agent to definitively know it matches the intent correctly when it encounters the same phrase again in future
-
the Validation tab allows you to review agent issues, intent issues and entity issues
Dialogflow can be integrated Telegram, Skype, Line, Twitter and Messenger from Facebook