This is a dedicated watch page for a single video.
A company wants to replace its manual process of reviewing claim compliance with an automated solution using Machine Learning. The company has a huge collection of claims stored in an S3 bucket. Every single claim under a compliance label comprises short sentences that have complex relationships with other claims. The Machine Learning Specialist assigned to this task prefers to use Amazon SageMaker built-in algorithms to train a supervised model that can classify if a claim is compliant or non-compliant. How can the Specialist create embeddings of these claims which can be fed as inputs for the downstream task?