AI/ML Based Intelligent Email Responder
Publication date: March 10, 2021 (Diagram history)
This architecture shows how to improve customer experience and agent productivity. With Amazon Simple Email Service (Amazon SES) and AWS ML services, you can comprehend intent and identify next-best-action for email responses.
AI/ML Based Intelligent Email Responder
The following steps describe the architecture:
-
Amazon SES captures emails received at the corporate email exchange.
-
Amazon SES writes the raw payload to Amazon Simple Storage Service and triggers a notification through Amazon Simple Notification Service. An AWS Lambda function captures this notification and triggers the email responder workflow.
-
The email responder workflow uses Amazon Comprehend for language, sentiment, and keyword extraction. It uses Amazon Translate for language translation. It calls intent extraction and next-best-action model endpoints deployed on Amazon SageMaker AI for better routing. It performs data lookups from the enterprise data lake. It updates the transactional data stores by using Amazon Kinesis Data Firehose.
-
Based on confidence and completeness, the workflow either sends a response through Amazon SES or routes the email to the right queue for an agent.
-
From the data collected, you continuously improve the intent extraction and next-best-action models by using SageMaker AI Ground Truth and SageMaker AI.
-
You can perform personalized email campaigns, additional email analytics, and reporting by using Amazon Pinpoint, Amazon Athena, and AWS Quick Sight.
Further reading
For additional information, refer to the following resources:
Diagram history
To be notified about updates to this reference architecture diagram, subscribe to the RSS feed.
| Change | Description | Date |
|---|---|---|
Initial publication | Reference architecture diagram first published. | March 10, 2021 |
Note
To subscribe to RSS updates, you must have an RSS plugin enabled for the browser you are using.
