Constructing Predictive Model for Subscription Fraud Detection Using Data Mining Techniques: The Case of Ethio-Telecom

No Thumbnail Available



Journal Title

Journal ISSN

Volume Title


Addis Ababa University


Access to telecommunications is critical to the development of all aspects of a nation‘s economy including manufacturing, banking, education, agriculture and government. However, the telecommunication services are not free from problem. Telecommunication fraud is the main problem of all telecom operators. Telecommunication fraud is the theft of telecommunication service (telephones, cell phones, computers etc.) or the use of telecommunication service to commit other forms of fraud. Victims include consumers, businesses and communication service providers. The subscription fraud is the most prevalent since with a stolen or manufactured identity, there is no need for a fraudster to tackle a digital network‘s encryption or authentication systems. This study is initiated with the aim of exploring the potential applicability of the data mining technology in developing models that can detect and predict pre paid mobile subscription fraud in Ethio-telecom service provision. The researcher selected around 25,000 records from six months collection of Call Detail Record data. After eliminating irrelevant and unnecessary data only a total of 21367 datasets are used for the purpose of conducting this study. The researcher also selected 14 attributes for this study based on their relevant for this research. The collected data has been preprocessed and prepared in a format suitable for the DM tasks. The study was conducted using WEKA software version 3.7.9 and four classification techniques namely J48, PART, Random forest and Multilayer perceptron of artificial neural network. As a result the Random forest algorithm registered better performance of 99.9251% accuracy running with 10-fold cross validation and used 14 attribute for this experimentation of this research. Future works are also implicated in this work.



Constructing Predictive Model