Technology news and Jobs 
Information Technology News Researchers seek to automate fraud detection
Information Technology News Researchers seek to automate fraud detection | Researchers seek to automate fraud detection |
|
| by Stephen Withers | |
| Wednesday, 07 November 2007 | |
  
Featured Whitepaper
5 Best Practices for Smartphone Support
Processes such as invoicing and salary payments would be monitored for indications of fraud. For example an employee might replace a supplier's bank account with his or her own before an invoice is paid, and then change it back afterwards. While this could be detected by a subsequent audit, the idea is to catch it at the time. A catalogue of fraud patterns has already been assembled, and the next step is a series of interviews to determine organisations' current practices. "Australian studies have found that employees are most likely to be the source of fraudulent behaviour and activity," said Andrew Clark, a researcher at QUT's Information Security Institute. "While many current strategies for dealing with fraud involve limiting users' capabilities through access controls, this project takes a different approach by using algorithms to analyse user behaviour for activities that might indicate that fraud is occurring." According to KPMG, Australian organisations reported frauds in excess of $150 million during 2006. "SMEs are highly impacted by fraud. Being smaller organisations, they often suffer more from a single fraud case, even it if it is just of average size, compared to large enterprises which have a stronger financial position in absolute terms," said Karsten Schulz, vice president of SAP Research Asia Pacific Japan. "The aim of the research is to extend fraud detection capabilities within SAP's SME solution offerings... It is a cat and mouse game because once fraudsters know you're looking for particular behaviour, they will change their behaviour in order to avoid detection." SAP's adoption of a Service Oriented Architecture allows functions such as fraud detection to be added to existing software. The SAP/QUT project is funded by an Australian Research Council Linkage Grant. |
Tags
| < Next story in category | Previous story in the category > |
|---|
