One-class näive bayesian classifier for toll fraud detection

Research output: Contribution to journalArticle

Abstract

In this paper, a one-class Naive Bayesian classifier (One-NB) for detecting toll frauds in a VoIP service is proposed. Since toll frauds occur irregularly and their patterns are too diverse to be generalized as one class, conventional binary-class classification is not effective for toll fraud detection. In addition, conventional novelty detection algorithms have struggled with optimizing their parameters to achieve a stable detection performance. In order to resolve the above limitations, the original Naive Bayesian classifier is modified to handle the novelty detection problem. In addition, a genetic algorithm (GA) is employed to increase efficiency by selecting significant variables. In order to verify the performance of One-NB, comparative experiments using five well-known novelty detectors and three binary classifiers are conducted over real call data records (CDRs) provided by a Korean VoIP service company. The experimental results show that One-NB detects toll frauds more accurately than other novelty detectors and binary classifiers when the toll frauds rates are relatively low. In addition, The performance of One-NB is found to be more stable than the benchmark methods since no parameter optimization is required for One-NB.

Original languageEnglish
Pages (from-to)1353-1357
Number of pages5
JournalIEICE Transactions on Information and Systems
VolumeE96-D
Issue number5
DOIs
Publication statusPublished - 2014 Jan 1
Externally publishedYes

Fingerprint

Classifiers
Detectors
Genetic algorithms
Industry
Experiments

Keywords

  • Genetic algorithm
  • Novelty detection
  • One-class Naive Bayesian classifier
  • Toll fraud detection

ASJC Scopus subject areas

  • Software
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence
  • Electrical and Electronic Engineering

Cite this

One-class näive bayesian classifier for toll fraud detection. / Kang, Pilsung.

In: IEICE Transactions on Information and Systems, Vol. E96-D, No. 5, 01.01.2014, p. 1353-1357.

Research output: Contribution to journalArticle

@article{c49d5ff97f8749b59636a9eea6e6f8c6,
title = "One-class n{\"a}ive bayesian classifier for toll fraud detection",
abstract = "In this paper, a one-class Naive Bayesian classifier (One-NB) for detecting toll frauds in a VoIP service is proposed. Since toll frauds occur irregularly and their patterns are too diverse to be generalized as one class, conventional binary-class classification is not effective for toll fraud detection. In addition, conventional novelty detection algorithms have struggled with optimizing their parameters to achieve a stable detection performance. In order to resolve the above limitations, the original Naive Bayesian classifier is modified to handle the novelty detection problem. In addition, a genetic algorithm (GA) is employed to increase efficiency by selecting significant variables. In order to verify the performance of One-NB, comparative experiments using five well-known novelty detectors and three binary classifiers are conducted over real call data records (CDRs) provided by a Korean VoIP service company. The experimental results show that One-NB detects toll frauds more accurately than other novelty detectors and binary classifiers when the toll frauds rates are relatively low. In addition, The performance of One-NB is found to be more stable than the benchmark methods since no parameter optimization is required for One-NB.",
keywords = "Genetic algorithm, Novelty detection, One-class Naive Bayesian classifier, Toll fraud detection",
author = "Pilsung Kang",
year = "2014",
month = "1",
day = "1",
doi = "10.1587/transinf.E97.D.1353",
language = "English",
volume = "E96-D",
pages = "1353--1357",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "5",

}

TY - JOUR

T1 - One-class näive bayesian classifier for toll fraud detection

AU - Kang, Pilsung

PY - 2014/1/1

Y1 - 2014/1/1

N2 - In this paper, a one-class Naive Bayesian classifier (One-NB) for detecting toll frauds in a VoIP service is proposed. Since toll frauds occur irregularly and their patterns are too diverse to be generalized as one class, conventional binary-class classification is not effective for toll fraud detection. In addition, conventional novelty detection algorithms have struggled with optimizing their parameters to achieve a stable detection performance. In order to resolve the above limitations, the original Naive Bayesian classifier is modified to handle the novelty detection problem. In addition, a genetic algorithm (GA) is employed to increase efficiency by selecting significant variables. In order to verify the performance of One-NB, comparative experiments using five well-known novelty detectors and three binary classifiers are conducted over real call data records (CDRs) provided by a Korean VoIP service company. The experimental results show that One-NB detects toll frauds more accurately than other novelty detectors and binary classifiers when the toll frauds rates are relatively low. In addition, The performance of One-NB is found to be more stable than the benchmark methods since no parameter optimization is required for One-NB.

AB - In this paper, a one-class Naive Bayesian classifier (One-NB) for detecting toll frauds in a VoIP service is proposed. Since toll frauds occur irregularly and their patterns are too diverse to be generalized as one class, conventional binary-class classification is not effective for toll fraud detection. In addition, conventional novelty detection algorithms have struggled with optimizing their parameters to achieve a stable detection performance. In order to resolve the above limitations, the original Naive Bayesian classifier is modified to handle the novelty detection problem. In addition, a genetic algorithm (GA) is employed to increase efficiency by selecting significant variables. In order to verify the performance of One-NB, comparative experiments using five well-known novelty detectors and three binary classifiers are conducted over real call data records (CDRs) provided by a Korean VoIP service company. The experimental results show that One-NB detects toll frauds more accurately than other novelty detectors and binary classifiers when the toll frauds rates are relatively low. In addition, The performance of One-NB is found to be more stable than the benchmark methods since no parameter optimization is required for One-NB.

KW - Genetic algorithm

KW - Novelty detection

KW - One-class Naive Bayesian classifier

KW - Toll fraud detection

UR - http://www.scopus.com/inward/record.url?scp=84900023990&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84900023990&partnerID=8YFLogxK

U2 - 10.1587/transinf.E97.D.1353

DO - 10.1587/transinf.E97.D.1353

M3 - Article

AN - SCOPUS:84900023990

VL - E96-D

SP - 1353

EP - 1357

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 5

ER -