The Chinese e-commerce company Alibaba is perhaps the biggest online commerce company in the world. Using its three main websites, Taobao.com, Tmall.com, and Alibaba.com, it hosts millions of businesses and hundreds of millions of users with $345 billion in 2018 sales last year (more than eBay and Amazon combined!). With so many transactions and so many users, Alibaba has worked to capture fraud signals directly from its extensive database of user behaviors and its network. It then analyzes them in real-time using machine learning to accurately sort the potentially fraudulent users from the good ones. Alibaba has developed five stages of fraud detection for each user:

1. account check,
2. device check,
3. activity check,
4. risk strategy, and
5. manual review.

These stages all combine to develop a risk score for each user. This fraud risk prevention score is so valuable to Alibaba and others that Alibaba shares and sells it to external customers.What will Data Analytics do next?