Unlike conventional systems that employ a uniform machine learning-based "detection" approach, FraudClear combines machine learning with direct access to airline reservation and loyalty systems to provide a fully integrated payment and loyalty fraud prevention platform.
This integration enables data extraction and analysis from Passenger Name Records (PNRs) directly at the source in the reservation system of the airlines without requiring any API modifications.
FraudClear facilitates complete automation within reservation systems, including manual review processes, exchange ticket workflows, NDC connections, access to check-in systems, and loyalty systems.
The fraud engine is highly customizable, employing a sophisticated algorithm that combines rule definitions, enhanced cross-referencing systems, and machine learning technology to align with the unique needs of each company.
Based on the scoring results, the fraud check application can trigger predefined actions on major Passenger Service Systems (PSS) and Global Distribution Systems (GDS). These actions may include cancelling reservations, suspending bookings, voiding transactions, processing refunds, inserting special remarks or executing specific queuing commands.
The fraud pattern recognition is adaptable to specific customer requirements and market conditions. Rules can be precisely defined using a multitude of metrics, such as IP geolocation, address normalization, device fingerprinting, email profiling, velocity checks, and internal and external data sources like negative and positive lists, customer solvency information, third party information sources such as Perseuss data, Ekata, Ethoca alerts, CPF numbers, and more.
