Fighting Fraud with Machine Learning at Shopify with Solmaz Shahalizadeh - TWiML Talk #60
Published:Oct 30, 2017 19:54
•1 min read
•Practical AI
Analysis
This article summarizes a podcast episode featuring Solmaz Shahalizadeh, Director of Merchant Services Algorithms at Shopify. The episode discusses Shopify's transition from a rules-based fraud detection system to a machine learning-based system. The conversation covers project scope definition, feature selection, model choices, and the use of PMML to integrate Python models with a Ruby-on-Rails web application. The podcast provides insights into practical applications of machine learning in combating fraud and improving merchant satisfaction, offering valuable lessons for developers and data scientists.
Key Takeaways
- •Shopify transitioned from a rules-based fraud detection system to a machine learning-based system.
- •The podcast discusses the importance of well-defined project scope and feature selection.
- •PMML is used to integrate Python machine learning models with a Ruby-on-Rails web application.
Reference
“Solmaz gave a great talk at the GPPC focused on her team’s experiences applying machine learning to fight fraud and improve merchant satisfaction.”