Use this URL to cite or link to this record in EThOS:
Title: A Policy Driven Approach to Proactive Fraud Management in Financial Data Streams
Author: Edge, Michael Edward
ISNI:       0000 0004 2693 5824
Awarding Body: The University of Manchester
Current Institution: University of Manchester
Date of Award: 2010
Availability of Full Text:
Access from EThOS:
Fraud management is a vital business operation for financial institutions towards minimising the widespread effect of fraud upon customer service delivery, bottom line operating expenditure and the organisation's brand image reputation. Rapidly changing fraud patterns continue to demonstrate fraudster's ability to actively reengineer their methods in response to ad-hoc security protocol deployments, and highlights the distinct gap between the speed of transaction execution within streaming financial data and corresponding fraud technology frameworks. Despite extensive research into fraud detection techniques using knowledge discovery techniques, a 'store now, query later' processing model simply no longer satisfies the fraud alerting requirements of multi-channel financial service platforms and financial institutions are migrating to increasingly proactive methods of fraud detection through real-time evaluation of streaming data channels prior to transaction completion. While this may reflect a simple shift of data processing from 'post' to 'pre' data storage, fraud strategy deployment continues to rely upon. the assembly of complex data processing architectures using low level application programming interfaces and existing solutions continue to address fraud detection upon a single channel, resulting in highly fragmented and disparate approaches to financial fraud management for complete service channel coverage. This thesis presents a policy based language and encompassing architecture framework for facilitating the conceptual level expression and implementation of proactive fraud controls within multi-channel financial service platforms. It is demonstrated how a domain specific language can be used to abstract the financial platform into a stream based information model to reduce implementation complexity and deployment latencies through an innovative. policy mapping architecture usable bY' both expert and non-expert users. Supporting components are developed and integrated into the framework model towards providing a comprehensive suite of assistive tools for underpinning a preventive and holistic approach to financial fraud management. Presented research provides important contributions related to policy driven system specification and maintenance within both financial fraud management, and also the wider information systems domain.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available