Repository logo
  • English
  • Català
  • Čeština
  • Deutsch
  • Español
  • Français
  • Gàidhlig
  • Italiano
  • Latviešu
  • Magyar
  • Nederlands
  • Polski
  • Português
  • Português do Brasil
  • Srpski (lat)
  • Suomi
  • Svenska
  • Türkçe
  • Tiếng Việt
  • Қазақ
  • বাংলা
  • हिंदी
  • Ελληνικά
  • Српски
  • Yкраї́нська
  • Log In
    New user? Click here to register. Have you forgotten your password?
Repository logo
  • Colleges, Institutes & Collections
  • Browse AAU-ETD
  • English
  • Català
  • Čeština
  • Deutsch
  • Español
  • Français
  • Gàidhlig
  • Italiano
  • Latviešu
  • Magyar
  • Nederlands
  • Polski
  • Português
  • Português do Brasil
  • Srpski (lat)
  • Suomi
  • Svenska
  • Türkçe
  • Tiếng Việt
  • Қазақ
  • বাংলা
  • हिंदी
  • Ελληνικά
  • Српски
  • Yкраї́нська
  • Log In
    New user? Click here to register. Have you forgotten your password?
  1. Home
  2. Browse by Author

Browsing by Author "Worku, Tarikua"

Now showing 1 - 1 of 1
Results Per Page
Sort Options
  • No Thumbnail Available
    Item
    Predictive Modeling for International Roaming Fraud Detection in Ethio Telecom
    (Addis Ababa University, 2018-03-04) Worku, Tarikua; Teferi, Dereje (PhD)
    Telecommunication fraud is remaining a challenging task since the beginning of commercial telecom service. There are various reasons that makes telecom fraud detection and prevention challenging. Integration of new technologies with ex-isting technologies without evaluating the security hole is main reason. Interna-tional roaming service is one of the immerged service in mobile technology. Roaming service allow subscribers to continue to use their home operator phone number, and other services while they are in another country. This geographical difference between service providers and subscribers make the roaming service open for different types of fraud attacks. So prevention and detection of interna-tional roaming fraud is crucial for service providers. In this study an effort has been made to build a predictive modeling for fraud detection using classification method. From decision tree and rule based classi-fication algorithms: random forest, J48, and ZeroR are used. Random Forest meet the highest accuracy 99.8154% with around 0.0109% false positive rate. So Ran-dom Forest algorithm is proposed as the best algorithm to detect international roaming fraud than the other two algorithms (J48 and ZeroR).

Home |Privacy policy |End User Agreement |Send Feedback |Library Website

Addis Ababa University © 2023