DATANAYA
Datanaya Levels-up Fraud Capability for Banks and Retailers

The Problem
Fraud is an expensive and complicated problem that requires constant attention. Tactics change rapidly and businesses need to keep up. As crimes become more sophisticated, so too must data analysts' tools.

Effective fraud protection needs to show connections – between people, accounts, transactions, and dates. Link analysis provides an effective solution.

When detecting fraud, analysts look for two types: known and unknown. Known fraud is fraudulent activity that's been encountered before, whereas unknown fraud hasn't been previously encountered.

The Solution
Link analysis plays an investigative role, providing fast and accurate situational analysis to inform a go/no-go decision in known fraud cases. Behaviour patterns can be defined and detected with rule-scoring and pattern-matching ('pattern of life') algorithms.

In unknown fraud cases, automated processes won't help. For these, link analysis acts as a fraud detection tool.

How it looks in action
Datanaya applied a case-centric approach to investigate known insurance fraud. Starting from a specific transaction, account or person, we worked outward to detect patterns.

To handle unknown fraud investigations, we applied a more global approach that loads large volumes of transactions into a chart to uncover outliers or unusual patterns.

Read our blog about insurance fraud to learn about this approach in more detail.