Fraud Detection with WSO2 Data Analytics Server

The WSO2 Fraud Detection Solution uses batch, real-time, predictive, and interactive analytics capabilities of WSO2 Data Analytics Server to convert domain knowledge into generic rules, implement fraud scoring, utilize Markov models and data clustering to model unknown types of fraud, and obtain interactive data querying and visualizations.

Diagram

Detecting fraud via known patterns using generic rules

The definition of ‘fraudulent behavior’ differs from one organization to another. Therefore, It is important to allow businesses to define what they perceive as fraud. WSO2 Fraud Detection Solution enables businesses to convert expert knowledge in their domain into a set of fraud rules. All transactions, individually and collectively, will then be compared against these fraud rules in real time and get flagged as fraud once a rule is violated.

Detecting unknown types of fraud via machine learning

WSO2 Fraud Detection Solution employs machine learning algorithms to model the ‘normal’ behavior of events and hence detect deviations from the modelled ‘normal’ behavior in real time data. We use ‘clustering’ mechanisms, which allow the modelling of ‘normal’ behavior as clusters, and anomalies (fraud) as deviations from those clusters.

Diagram

Detecting rare activity sequences using Markov Modeling

Some advanced fraud techniques involve a sequence of transactions that individually look legitimate, but taken together as a sequence, form an instance of fraud. We employ Markov Models to calculate event sequence probabilities, and flag incoming activity sequences as fraudulent if they have an extremely low probability of occurring in sequence. This enables enterprises to detect advanced fraud techniques employed by international fraud rings and organized criminal networks.

Reduce false alarms using fraud scoring

With a strong fraud detection system in place enterprises face the danger of losing customers to overprotective fraud rules. WSO2 Fraud Detection Solution utilizes scoring mechanisms to solve this problem. It enables organizations to use a combination of rules with a weight attached to each rule, and generates a single number that reflects how well a transaction performed against multiple fraud indicators. This single number is then checked against a threshold, which will trigger focused alerts for fraud and fraud only.

Diagram

Caught them in the act - what next?

Once a possible fraud is detected, an enterprise must dig deeper to understand whether there are any other events/relationships that this event is linked to. It pays to query other events that have something in common with the event that was flagged as possible fraud.

The WSO2 Fraud Detection Solution provides useful visualizations and further querying facilities for a fraud analyst to discover possible relationships within events, enabling the discovery of large fraud rings and collusions.

Diagram

Interested?

Read white paper on ‘Fraud Detection and Prevention - A Data Analytics Approach’ to learn more about the use of analytics to detect and prevent fraud.