[ASK] HowTo catch some Fraud Scenarios in ecommerce – INTERVIEW Q
1 comments
Here are some which I thought... I am wondering if there are other classes or categories we should watch for Fraud/Risk
1) Unusually large orders or high-priced orders 2) Expedited shipping on large quantities or high-priced orders 3) Expedited shipping when billing and shipping addresses differ 4) Make sure the billing address matches the IP location. 5) Limit the number of declined transactions. 6) Customers who make multiple orders from different credit cards 7) Machine Learning - supervised classification - the system is told in advance which already-reviewed orders are fraud, and then asked to predict whether another order is fraud based on what it’s seen so far 8) Anomaly detection
1) Unusually large orders or high-priced orders 2) Expedited shipping on large quantities or high-priced orders 3) Expedited shipping when billing and shipping addresses differ 4) Make sure the billing address matches the IP location. 5) Limit the number of declined transactions. 6) Customers who make multiple orders from different credit cards 7) Machine Learning - supervised classification - the system is told in advance which already-reviewed orders are fraud, and then asked to predict whether another order is fraud based on what it’s seen so far 8) Anomaly detection
this is interview question and I will like to brainstrom with you guys. Lets say you are interviewing for ecommerce company, what are the things you will look to catch fraud scenarios.
Lets assume there are following things in this system: 1) Buyer 2) Seller 3) Recommendations for Buyer and Seller (buyer rate Seller and Seller rates buyer) 4) Order Processing 5) Payment 6) Purchase
Again this is just open ended question and skys the limit for the solution. So, wanted to see how many ways we can catch fraud scenarios. This may/may not involve any machine learning algorithms.
Cheers!