What should the company do to reduce the processing time of loan applications?
Configure Amazon Textract to route low-confidence predictions to Amazon SageMaker Ground Truth. Perform a manual review on those words before performing a business validation.
Use an Amazon Textract synchronous operation instead of an asynchronous operation.
Configure Amazon Textract to route low-confidence predictions to Amazon Augmented AI (Amazon A2I). Perform a manual review on those words before performing a business validation.
Use Amazon Rekognition’s feature to detect text in an image to extract the data from scanned images. Use this information to process the loan applications.
Explanations:
Amazon SageMaker Ground Truth is used for building custom machine learning models and labeling data, but it is not ideal for manual review of low-confidence predictions from Textract. This does not directly address improving the validation process in loan applications.
Synchronous operations in Amazon Textract are better for low-volume, real-time processing. However, for processing thousands of documents, asynchronous operations are more scalable. This change would not necessarily reduce the processing time.
Amazon Augmented AI (A2I) allows for human review of low-confidence predictions in Amazon Textract, streamlining the manual review process for documents with errors. This helps reduce the time spent on business validation by focusing human review only on difficult cases.
Amazon Rekognition’s text detection is not specifically designed for extracting structured data from scanned legal documents. Textract is optimized for this type of structured text extraction, making Rekognition less suitable for loan application processing.
From my perspective, the answer is:
Configure Amazon Textract to route low-confidence predictions to Amazon Augmented AI (Amazon A2I). Perform a manual review on those words before performing a business validation.