Credit risk prediction with Azure Machine Learning

Credit risk analysis is important to financial institutions that provide loans to businesses and individuals. Credit loans and finances have risk of being defaulted or delinquent. To understand risk levels of credit users, credit providers normally collect vast amount of information on borrowers. Statistical predictive analytic techniques can be used to analyze and determine risk levels involved on credits, and approve or reject credit applications accordingly.

My workshop at Directions EMEA 2017 (Credit Risk Prediction with Azure Machine Learning) demonstrates how to perform cost-sensitive binary classification in Azure Machine Learning to predict credit risk based on the information given on a credit application built in a Dynamics 365 solution.

We will start by practicing with Azure Machine Learning Studio to understand its basic concepts. Machine learning is a data science technique that allows computers to use existing data to forecast future behaviors, outcomes, and trends. Using machine learning, computers learn without being explicitly programmed. This is what Azure ML Studio does for us: it allows non-data scientists and computer programmers to have easy access to data mining processes, with the purpose of predicting outcomes by looking at patterns in existing datasets. And the outcome is exposed as a web service that can be consumed over HTTPS as a REST API by any third-party application, including Dynamics 365.

 Azure Machine Learning Basic Workflow. (Image courtesy of Microsoft.)

Walking through the next implementation steps, we will define our initial dataset by accessing publicly available credit risk data, and then develop and train a predictive model based on that dataset. Then ...