View a markdown version of this page

Guidance for Contextual Intelligence for Advertising - Guidance for Contextual Intelligence for Advertising
Architecture diagramFurther readingDiagram history

Guidance for Contextual Intelligence for Advertising

Publication date: April 6, 2022 (Diagram history)

With this architecture, you can build a contextual advertising solution that uses machine learning to reach target audiences without third-party cookies. Contextual advertising reaches an audience based on the content that users consume. The solution uses an event-driven, serverless architecture that is scalable and cost-optimized.

This architecture helps demand-side platforms (DSPs), advertisement publishers, and supply-side platforms (SSPs) build a contextual intelligence solution. You use AWS AI and machine learning services to extract relevant metadata and map it to your own or an industry-standard taxonomy.

Architecture diagram

Content flowing through AWS serverless services for analysis, taxonomy mapping, and programmatic advertising bid enrichment.

The following steps describe the architecture:

  1. DSPs, SSPs, or ad publishers invoke an API on Amazon API Gateway to trigger content discovery. The API fetches text, images, audio, and video from the provided content.

  2. The serverless crawler is built on AWS Step Functions to orchestrate exploration and download of content. Amazon DynamoDB stores ephemeral discovery data.

  3. You can use a crawler or a content management system API to extract media types and store them in an Amazon S3 bucket.

  4. Amazon S3 stores content (text, image, video), with available metadata in Amazon DynamoDB for analysis.

  5. A content discovery completion event starts controller orchestration built on AWS Step Functions, AWS Lambda, and Amazon SNS.

  6. Amazon SNS events invoke AWS Lambda functions to start content analysis by using Amazon Comprehend, Amazon Rekognition, and Amazon Transcribe.

  7. Amazon DynamoDB stores topics, sentiment, and object labels from the content analysis workflow.

  8. The Contextual Intelligence Taxonomy Mapper (CITM) uses the Bidirectional Encoder Representations from Transformers (BERT) model deployed on Amazon SageMaker AI. CITM maps metadata in Amazon DynamoDB to an industry-standard taxonomy such as the IAB Content Taxonomy.

  9. A AWS Lambda function gets the mapping and stores it in Amazon DynamoDB within the category service.

  10. Bidding servers invoke an API built on Amazon API Gateway to fetch categories from Amazon DynamoDB. This informs programmatic advertising bids with low latency.

Further reading

For additional information, see the following resources:

Diagram history

To receive updates about this reference architecture diagram, subscribe to the RSS feed.

ChangeDescriptionDate

Initial publication

Reference architecture diagram first published.

April 6, 2022

RSS subscription requirement

To subscribe to RSS updates, you must have an RSS plugin enabled for the browser you are using.