Fraud prediction using AutoAI

Summary

This code pattern discusses building a system for creating predictions that can be used in different scenarios. It focuses on predicting fraudulent transactions, which can reduce monetary loss and risk mitigation. But, you can use the same approach for predicting customer churn, demand and supply forecast, and more.

Description

Automation and artificial intelligence (AI) technology is transforming businesses. They address challenges in areas of healthcare, technology, and more. At the same time, these technologies transform the nature of work and the workplace itself. In this code pattern, we focus on building systems for churning out predictions that can be used in different scenarios. We try to predict fraudulent transactions that we know can reduce monetary loss and risk mitigation. The same approach can be used for other scenarios like predicting customer churn or demand and supply forecast. Building predictive models requires time, effort, and knowledge of algorithms to create effective systems that can predict the outcome accurately. With that being said, IBM has introduced AutoAI, which automates all of the tasks involved in building predictive models for different requirements. This code pattern shows how AutoAI can churn out great models quickly, which saves time and effort and aids in a faster decision-making process. AutoAI can be run in public clouds and in private clouds, including IBM Cloud Pak™ for Data.

When you have completed this code pattern, you understand how to:

  • Quickly set up the model building services on IBM Cloud
  • Ingest the data and initiate the AutoAI tool
  • Build different models using AutoAI and evaluate the performance
  • Choose the best model and complete the deployment
  • Generate predictions using the deployed model by making REST calls
  • Compare the process of using AutoAI and building the model manually

Flow

Fraud prediction AutoAI

  1. Log in to Watson Studio, create a project, and initiate an instance of AutoAI and Cloud Object Storage.
  2. Upload the .csv data file to Object Storage.
  3. Initiate the model building process using AutoAI and create pipelines.
  4. Evaluate different pipelines from AutoAI and select the best model for deployment.
  5. Generate accurate predictions by making REST calls to the deployed model.

Instructions

Find the detailed steps for this pattern in the README. Those steps show you how to:

  1. Create an account with IBM Cloud.
  2. Create a new Watson Studio project.
  3. Add data to the project.
  4. Add an asset as AutoAI.
  5. Create and define an experiment.
  6. Import the .csv file.
  7. Run the experiment.
  8. Analyze the results.
  9. Deploy to IBM Cloud.
  10. Test the model.
Sharath Kumar RK
Manjula Hosurmath