Machine Learning is being applied to a variety of use cases including fraud prevention, anti-money laundering (AML) and eCommerce product recommendation. As you apply machine learning to identify anomalous behavior such as finding fraudsters or money launderers, it is akin to finding needles in a massive haystack - companies must sort and make sense of massive amounts on data in order to find the "needles" or in this case, the fraudsters.
In this paper we review a use case on how TigerGraph can train the machine for fraud detection with graph based features. This in turn makes the machine smarter and more successful in recognizing potential scam artists and fraudsters, enabling companies to find that needle in the haystack.