A food-service company wants an ML model to reduce daily food waste and boost sales, and it needs to continually improve model accuracy over time. Which solution meets these needs?

Amazon SageMaker is a fully managed service for building, training, and deploying machine learning models. Its capabilities allow for iterative retraining with new data, which is crucial for a model that needs to continually improve accuracy as new food waste and sales data become available. This approach directly addresses the requirement for ongoing model enhancement. Amazon Personalize is designed for real-time personalization and recommendations, not general-purpose predictive modeling for waste reduction. Amazon CloudWatch is a monitoring service, not an ML platform for model development or deployment. Amazon Rekognition is a computer vision service, unsuitable for analyzing food waste or sales data.