Akademik Veri Yönetim Sistemi
3rd International Conference on Electrical, Communication and Computer Engineering, ICECCE 2021, Kuala-Lumpur, Malezya, 12 - 13 Haziran 2021
© 2021 IEEE.In the telecommunication industry, the prediction of customer churn behavior is a subject of active research. Features derived from customers' use of telecom infrastructure are often used to predict customer churn behavior. However, the complex networks created by the communication data between the customers and the features to be obtained from these networks can also affect customer churn behavior. Within the scope of this research, features, which are used to predict customer churn behavior by using Social Network Analysis (SNA) techniques on complex networks formed as a result of customer interaction on telecom infrastructures, are proposed. In addition to that, a data analysis workflow method that can predict customer churn behavior is suggested. A prototype application of the proposed method was developed and its success in predicting customer churn behavior was evaluated with experimental studies. In this study, an anonymized data set belonging to a telecom industry firm is used. The results obtained show that the proposed method can make successful predictions and is usable.