Preventing Fraud and Saving Costs with Machine Learning - Dataconomy

Whilst most businesses don't earn revenue by processing data, they do spend a large amount of their hard earned revenue in manually processing data, validating it and ultimately performing manual tasks that don't scale. But at what point does this manual involvement become a burden of cost upon your business? And really how much manual involvement should be required? Take payment fraud for example. According to the 2015 Merchant Risk Council (MRC) Global Fraud Survey, merchants typically manually review 10-15% of online orders.