Scalable Fraud Detection With Multi-Model
Across industries, fraud is a growing problem resulting in a global annual loss of $3.7 trillion. Fraudsters became more sophisticated in hiding their activities by forming fraud rings, using stolen identities and other patterns. Traditional approaches still focus on discrete data missing many opportunities to identify or prevent fraud.
Multi-model lets organizations see data from different perspectives, its context and detect fraud patterns with graph database technology even within large scale datasets. In this white paper we will show how to convert data from relational to multi-model graphs, how various fraud detection queries work in ArangoDB’s Query Language (AQL) and how this Fraud Detection can be done at scale.