Real Anomaly Detection in Telecommunication Multidimensional Data Using Data Mining Techniques

Abstract

The tremendous amount of data are generated and collected by telecommunication companies. These data include call detail data which describe the calls traversing the telecommunication networks as well as network and customer data which mainly describe incomes of telecommunication companies. The amount of data is so huge that manual analysis of these data is impossible. The need to automatically handle such large volumes of data has led to the development of special algorithms and technologies such as data mining, intelligent computer agents, knowledge-based expert systems, etc. Telecommunication companies are strongly interested not only in identifying fraudulent phone calls and identifying network faults but also in forecasting the preferred directions of customer calls or the incomes of the companies. The paper presents a communication real anomaly detection framework, which uses data mining technologies using OLAP cube built for telecommunication data.