Një model hibrid për eksplorimin e të dhënave dhe përcaktimi i planeve tarifore për abonentët e telefonive celulare
A hybrid model for data exploration and determination of tariff plans for mobile phone subscribers

Summary/Abstract: Nowadays, customer churn has become a serious risk to business success. Mobile telecommunications market is particularly affected by this phenomenon. Due to very rapid development of technology, tele-communication market is changing dynamically. This dynamism and the daily necessity of communica-tion imply constant change of preferences and requirements of customers. Consequently, telecom mo-bile customers are likely to change often the companies from which they choose to receive services. Due to this phenomenon, these companies are constantly under risk of customer churn. One of the reasons of customer churn in telecommunications is selection of the wrong tariff plan by the customers. Therefore, the aim of this study is to develop a hybrid data mining model, which can support customer behavior analysis in order to find the appropriate tariff plan for each customer. In this way, customer portfolio management will be more efficient and customer churn will be prevented. To develop this model we will utilize data mining techniques of clustering and classification. By compar-ison we will evaluate the hybrid models efficiency in comparison to benchmark cases, when no combi-nation of data mining techniques is applied. The model will be tested on Rapid Miner tool and will be applied on real datasets.

• Details
• Contents