A Financial Services Company leverages Machine Learning to Secure Data Processing, Improved Fraud Detection, and Smart Decision-Making

Solutions

• The dataset uploaded to Amazon S3 triggers Event Bridge for task triggers in the pipeline.  
• The XML dataset was parsed using Amazon SageMaker Pipeline with the XML parsing step. 
• Amazon SageMaker Pipelines manage the machine learning lifecycle, Amazon SageMaker Processing steps used for feature engineering, and EDA and Amazon SageMaker Training steps used for real-time and batch model training. 
• Trained model artifacts are automatically stored in Amazon S3 and registered in the model registry. 
• Amazon SageMaker Evaluation generates metrics for classification and regression. 
• The approved model was deployed in real-time applications using Amazon SageMaker Endpoint. 
• Architecture automated with AWS CloudFormation templates and CI/CD tools for pipeline rebuilding. 
• CI/CD pipeline with 2 Code Pipelines integrating GitHub and AWS CodeCommit repositories.