Offer Decisioning with Machine Learning


I was given a dataset of 3 tables with transaction logs, customer attributes, and offer attributes. I have:

  • built a preprocessing program to transform this data to prepare for modelling (as shown here)
  • tested many different Machine Learning models, exported the best one in a pickle file,
  • built a web application with Flask that takes customer attributes as inputs from a web form and uses this ML model to find the best marketing offer,
  • built a Continuous Deployment pipeline that automatically deploys the latest version of this web app from my GitHub repository to my server.