Non-obvious indicators that a transaction might be fraudulent

Re:Non-obvious indicators that a transaction might be fraudu... by cheez on Tuesday April 05, 2016 @11:06PM (#11436336)

They call it machine learning, so did they apply unsupervised learning to determine these factors? How did they determine these factors were relevant?

[ Parent | on HN ]

Re:Non-obvious indicators that a transaction might be fraudu... by nickcano on Wednesday April 06, 2016 @02:48PM (#11440845)

They could have used PCA[1] on their feature set and let a human come up with these; they claim to use machine learning, not that this checklist was generated as a result of machine learning.

Otherwise, maybe they have an algorithm that looked at resulting classifications from learning and flagged inputs with the largest and most consistent variance across classes.

[1]https://en.wikipedia.org/wiki/Principal_component_analysis

[ Parent | on HN ]

Re:Non-obvious indicators that a transaction might be fraudu... by cheez on Wednesday April 06, 2016 @06:07PM (#11442563)

I'm OK with a human being involved, I'm curious how they actually figured it out.

[ Parent | on HN ]