Toward a new cloud-based approach to preserve the privacy for detecting suspicious cases of money laundering in an investment bank
|Title:||Toward a new cloud-based approach to preserve the privacy for detecting suspicious cases of money laundering in an investment bank||Authors:||Le-Khac, Nhien-An
|Permanent link:||http://hdl.handle.net/10197/6553||Date:||12-Jun-2014||Online since:||2015-05-13T11:54:01Z||Abstract:||Today, money laundering poses a serious threat not only to financialinstitutions but also to the nations. This criminal activity is becoming more andmore sophisticated and seems to have moved from the clich of drug traffickingto financing terrorism and surely not forgetting personal gain. Mostinternational financial institutions have been implementing anti-moneylaundering solutions to fight investment fraud. On the other hand, cloud-basedapplications are merging daily and bringing to clients with lower cost ofplatforms and data storage, greater scalability and improved businesscontinuity. Hence, more financial instituitions aim to move their ITinfrastructure to the cloud. However, accessing directly to the customertransaction datasets by a third party could be a confidential issue. This approachis more severe when these solutions are built by collaborating partners.Traditional methods are based on data access agreement but there is still a riskof infringing privacy. In order to preserve the privacy of datasets, different datadisguising methods have been proposed. Nevertheless, analysing disguiseddatasets is a performance issue in the context of detecting suspicious moneylaundering cases where the real value of data has an important impact. Indeed,the results of analysis could also be a privacy issue. Within the scope of acollaboration project for developing a new cloud-based solution for the Anti-Money Laundering Units in an international investment bank, in this paper, wepropose new cloud-based approach using data disguising methods applied inanalysing transaction datasets. We also show that the creating relevantdimensions from the current ones is efficient for analysing transaction datasetsin terms of both detecting suspicious case and privacy preserving.||Funding Details:||Science Foundation Ireland||Type of material:||Conference Publication||Keywords:||Machine learning; Statistics||Other versions:||http://www.iccs-meeting.org/iccs2014/||Language:||en||Status of Item:||Peer reviewed||Conference Details:||International Conference on Computational Science, (ICCS 2014), Cairns, Australia, 10-12 June 2014|
|Appears in Collections:||Computer Science Research Collection|
Insight Research Collection
Show full item record
Page view(s) 5030
This item is available under the Attribution-NonCommercial-NoDerivs 3.0 Ireland. No item may be reproduced for commercial purposes. For other possible restrictions on use please refer to the publisher's URL where this is made available, or to notes contained in the item itself. Other terms may apply.