Machine learning (ML) is a powerful tool that can help organizations increase revenue, drive business growth, and reduce costs by optimizing core business functions such as supply and demand forecasting, customer churn prediction, credit risk scoring, pricing, predicting late shipments, and many others.

Conventional ML development cycles can be time-consuming and require a deep understanding of data science and ML development skills. However, in this post, we will explore how a financial or business analyst at a bank can easily predict a customer’s loan status without needing to become a machine learning expert.

Business Use Case: Banking Institution

Let’s dive into a business use case for a banking institution. We will demonstrate how a financial or business analyst can predict if a customer’s loan will be fully paid, charged off, or current using a machine learning model. The process will involve pulling in the necessary data, cleaning it up, and building a predictive model without requiring intricate ML knowledge.

Services Used

Our solution leverages Amazon SageMaker Canvas, a web-based visual interface for building, testing, and deploying machine learning workflows. The integration with Data Wrangler allows for easy data preparation and analysis using natural language commands. Additionally, Amazon Redshift, a petabyte-scale data warehouse, provides powerful data analytics capabilities.

Solution Architecture

• A business analyst signs in to SageMaker Canvas.
• The analyst connects to the Amazon Redshift data warehouse to pull the required data.
• SageMaker Canvas builds a predictive analysis ML model.
• Batch prediction results are obtained from the model.
• The results are sent to Amazon QuickSight for further analysis.

Implementation Steps

In the implementation steps, data is loaded into the Amazon Redshift cluster, and the chat for data option in SageMaker Canvas is set up to allow for data preparation using natural language commands. The model is then created, and predictions are generated, which can be further analyzed using Amazon QuickSight.

Analyze the Predictions

The predictions from SageMaker Canvas can be visualized in QuickSight to gain valuable insights from the data. QuickSight provides a range of interactive visualization options and allows for sharing dashboards with other users within the organization.

Conclusion

By integrating Amazon Redshift with SageMaker Canvas, organizations can accelerate ML solutions without the need for extensive data movement or ML experience. This enables business analysts to derive valuable insights from data while data scientists and ML engineers can refine and extend models as required.

About the Authors

• Suresh Patnam – Principal Sales Specialist AI/ML at AWS
• Sohaib Katariwala – Sr. Specialist Solutions Architect at AWS
• Michael Hamilton – Analytics & AI Specialist Solutions Architect at AWS
• Nabil Ezzarhouni – AI/ML and Generative AI Solutions Architect at AWS