Fraud & Risk
How to Choose a Machine Learning Platform for Risk
Even as an increasing number of eCommerce merchants use machine learning as part of their risk mitigation strategy, it can be challenging to objectively determine which platform best meets their needs and goals. This paper describes three impacts which poor fraud detection has on online businesses, followed by seven criteria which may be used to weigh the pros and cons of each system. Particular emphasis is given to Random Forest and Neural Network algorithms as well as to entity profiles / behavioral analytics. A checklist summarizing the seven criteria and a brief overview of Feedzai conclude the whitepaper.