Classification of SIM Box Fraud Detection Using Support Vector Machine and Artificial Neural Network


  • Roselina Sallehuddin Universiti Teknologi Malaysia
  • Subariah Ibrahim Universiti Teknologi Malaysia
  • azlan mohd zain Universiti Teknologi Malaysia
  • Abdikarim Hussein Elmi Universiti Teknologi Malaysia



SIM box fraud is classified as one of the dominant types of fraud instead of subscription and superimposed types of fraud. This fraud activity has been increasing dramatically each year due to the new modern technologies and the global superhighways of communication, resulting the decreasing of the revenue and quality of service in telecommunication providers especially in Africa and Asia. This paper outlines the Artificial Neural Network (ANN) and Support Vector Machine (SVM) to detect Global System for Mobile communication (GSM) gateway bypass in SIM Box fraud. The suitable features of data obtained from the extraction process of Customer Database Record (CDR) are used for classification in the development of ANN and SVM models. The performance of ANN is compared with SVM to find which model gives the best performance. From the experiments, it is found that SVM model gives higher accuracy compared to ANN by giving the classification accuracy of 99.06% compared with ANN model, 98.71% accuracy.

Author Biography

Roselina Sallehuddin, Universiti Teknologi Malaysia


Name:          Roselina binti Salleh @ Sallehuddin
Address:       Faculty of Computer Science and Information System
University Teknologi Malaysia
Tel:             07-5532082
Email: This e-mail address is being protected from spambots. You need JavaScript enabled to view it


May 1988        Diploma Science Computer (Second Class)
Universiti Teknologi Malaysia
May 1991        Bachelor of Science Computer (Second Class Upper)
Universiti Teknologi Malaysia
August 1999        Master of Science (Computer Science)
Universiti Teknologi Malaysia
Thesis Title: Penggunaan Rangkaian Neural dalam Peramalan Data Siri Masa Bermusim
9 July 2010        PhD (Computer Science)
University Technology Malaysia
Thesis Title:     Hybridization of Nonlinear and Linear Models for Time Series Forecasting


June- Sept, 1990        System Analyst, Jabatan Perkhidmatan Awam
(Skim Khidmat Sementara)
Sept1990-Nov, 1991    Tutor, Pusat Pengajian Sains Matematik & Sains Komputer, USM, P.Pinang.
Nov 91- August 99        Assistant Lecturer, Faculty Science Computer and Information System, UTM
August 99 – Dec 2009     Lecturer, Faculty Science Computer and Information System, UTM
14th  Jan – till present    Senior Lecturer, FSKSM


Soft Computing Techniques & Algorithms
Inteligent Data analysis and Classification, Modeling and Forecasting


Experience of Research:    11 years

Research/ Main Project Involved:

Research Title    Sponsor    Vote Number    Leader    Member    Begin    End
Ranking Critical factor with cooperative feature selection for time series    RMC    77316    Roselina Sallehuddin    Prof Dr Siti Mariyam    15 Dis 2009    14 Jun 2011
Three Term Neural Network Weight Adaptation for Global Optimum Solution    FRSG    78243    Razana Alwee    Roselina Sallehuddin P.M Dr. Siti Mariyam Hj. Shamsuddin    Sept, 2007    August, 2009
Development of Intelligent System For Predicting Stock Market Price Returns

MOSTE    74083
P.M Dr. Siti Mariyam Hj. Shamsuddin    Roselina Sallehuddin Razana Alwee    2002    2005
Developing Real Time Flood Forecasting
IRPA    74018    P.M. Dr. Salihin Ngadiman    Roselina SallehuddinRazana Alwee
Nurulhuda Firdaus    2002    2004
Development of Compound Clustering Techniques using Hybrid Soft computing Algorithm    IRPA    71760    P.M Dr. Naomie Salim    Roselina SallehuddinRazana Alwee, Mahadi Bahari, Aryati Bakri    2003    2005
(Sistem Maklumat Universiti)
IRPA        P.M Dr. Mohd Zailani
Roselina SallehuddinHalimah Hassan,Zuraini Ismail,Rohani Hassan    1994    1996
P.M. Noraniah Mohd. Yasin
Roselina SallehuddinNorhaizan Mohd.Radzi    2002    2004
Classification and Clustering of Deseasonalised Time Series Data using
soft computing techniques
UPP    71760    Roselina Sallehuddin    P.M Dr. Siti Mariyam Shamsuddin    2001    2002
A Study on the performance of backpropagation and individual
forecast technique in developing Real Time Flood forecasting model
for Johor Catchment Area
UPP    71834    Nurulhuda Firdaus    Roselina Sallehuddin, Shukor Talib,     2002    2003




How to Cite

Sallehuddin, R., Ibrahim, S., mohd zain, azlan, & Hussein Elmi, A. (2014). Classification of SIM Box Fraud Detection Using Support Vector Machine and Artificial Neural Network. International Journal of Innovative Computing, 4(2).