Development of users' call profiles using unsupervised random forest | NOVA. The University of Newcastle's Digital Repository

Home

Please use this identifier to cite or link to this item: http://hdl.handle.net/1959.13/919135

Title: Development of users' call profiles using unsupervised random forest
Author/Creator: Saaid, Fatimah Almah; King, Robert; Nur, Darfiana
Institution: The University of Newcastle. Faculty of Science & Information Technology, School of Mathematical and Physical Sciences
Description: The aim of this paper is to detect fraud in telecommunications data which consists of millions of call records generated each day. The fraud detection is implemented via the construction of user call profiles using the calls detail records (CDR) data. This paper attempts to investigate the reliability of the unsupervised Random Forest method in building the profiles using its variable importance measure. Four different simulation scenarios, using different number of variable selection in each node of the tree, are performed.
Relation: 3rd Annual ASEARC Research Conference. ASEARC: Proceedings of the Third Annual ASEARC Research Conference (Newcastle, N.S.W. 7-8 December, 2009)
Relation: http://www.uow.edu.au/informatics/maths/research/groups/asearc/2009rescon/index.html
Date: 2009
Publisher: Applied Statistics Education and Research Collaboration (ASEARC)
Keyword(s): telecommunications fraud; users' call profiles; unsupervised Random Forest; variable importance
Resource Type: conference paper
Identifier: http://hdl.handle.net/1959.13/919135
Identifier: ISBN:9780980603484

31 Visitors

33 Hits

NOVA. The University of Newcastle's Digital Repository