Consumer Review Sentiment Analysis for CPG
Publication date: May 13, 2021 (Diagram history)
With this architecture, you can gain insights from consumer reviews in near real time. Consumer packaged goods (CPG) companies use review data to understand consumption-related decisions and proactively address issues. You use Amazon Comprehend for natural language processing (NLP) sentiment analysis, Amazon Translate for translation, and a serverless event-driven architecture.
For more information about this approach, see Detect
sentiment from customer reviews using Amazon Comprehend
Sentiment analysis diagram
The following steps describe the architecture:
-
Review data flows into AWS through the ingestion pipeline from multiple sources such as ecommerce sites, product websites, and social media webhooks.
-
An Amazon Data Firehose delivery stream loads the reviews into an Amazon Simple Storage Service bucket.
-
Amazon AppFlow ingests data from software-as-a-service (SaaS) applications such as Salesforce, Marketo, Slack, and ServiceNow into Amazon S3.
-
An Amazon S3 event invokes an AWS Lambda function to analyze raw reviews. Use Amazon Translate to translate reviews into a base language if necessary. Use Amazon Comprehend for NLP to perform sentiment analysis, key-phrase extraction, and entity extraction.
-
The AWS Glue Data Catalog contains a logical database that organizes tables for the data in Amazon S3.
-
Amazon Athena uses these table definitions to query data in Amazon S3. Results display in an Amazon Quick Sight dashboard.
Further reading
For additional information, see the following resources:
Diagram history
To receive updates about this reference architecture diagram, subscribe to the RSS feed.
| Change | Description | Date |
|---|---|---|
Initial publication | Reference architecture diagram first published. | May 13, 2021 |
RSS subscription requirement
To subscribe to RSS updates, you must have an RSS plugin enabled for the browser you are using.