Customer support · Production
super.AI intent clustering improves customer service chatbot relevance for Facebook
The problem
Facebook's customer service chatbot needed to better determine user intent in order to serve more relevant help articles to customers.
Workflow diagram · grounded in source
1
User submits query
trigger
“A user types in a query to the bot”
2
Intent clustering
ai_action
“we clustered each article and provided an exclusive choice output Not helpful, Somewhat Helpful, Very Helpful”
3
Article served to user
output
“The bot sends the user an article based on what it thinks the intent is”
Reported outcome
Intent clustering provided by super.AI increased the relevance of the chatbot, improving overall customer service and reducing dependency from human agents.
Reported metrics
Chatbot relevanceincrease the relevance of its chatbot
Dependency from human agentsreducing dependency from human agents
Reported stack
super.AI
Frequently asked questions
What did this team achieve with this AI workflow?
Intent clustering provided by super.AI increased the relevance of the chatbot, improving overall customer service and reducing dependency from human agents.
What tools did this team use?
super.AI.
What results were reported?
Chatbot relevance: increase the relevance of its chatbot; Dependency from human agents: reducing dependency from human agents (source-reported, not independently verified).
How is this customer support AI workflow structured?
User submits query → Intent clustering → Article served to user.