Amazon Web Services764 тыс
Опубликовано 9 июля 2024, 13:52
Slalom built a solution on behalf of a client that guides regulation as it becomes law. Using generative AI, they designed a machine learning (ML) solution that accelerates the review of hundreds of thousands of comments made against policies on regulations.gov. Using AWS Lambda for daily ingestion of comments and S3 for data storage, comments are pulled into Amazon SageMaker for data cleansing and enrichment then forwarded to an ML workflow powered by Amazon Bedrock & Amazon Sagemaker. Amazon Sagemaker notebooks were used to access HuggingFace FLAN UL2 and scikit-learn machine learning library to understand sentiment plus data clustering and analysis. While Amazon Titan was used for expert designation and more. The results of this analysis are then delivered back to S3 which is used as a source for their client to consume the data with their business intelligence tool of choice. This solution provides a new capability to a small non-profit, allowing them to use a lean staff to process vast amounts of data that historically was a challenge to manually review.
Check out more resources for architecting in the #AWS cloud:
amzn.to/2ZIbygO
#AWS #AmazonWebServices #CloudComputing #ThisIsMyArchitecture #AmazonSagemaker #AmazonTitan #AmazonBedrock #GenAI #Lambda #GenerativeAI #AIML
Check out more resources for architecting in the #AWS cloud:
amzn.to/2ZIbygO
#AWS #AmazonWebServices #CloudComputing #ThisIsMyArchitecture #AmazonSagemaker #AmazonTitan #AmazonBedrock #GenAI #Lambda #GenerativeAI #AIML
Свежие видео