Applying data mining to telecom chum management

Taiwan deregulated its wireless telecommunication services in 1997. Fierce competition followed, and chum management becomes a major focus of mobile operators to retain subscribers via satisfying their needs under resource constraints. One of the challenges is chumer prediction. Through empirical evaluation, this study compares various data mining techniques that can assign a 'propensity-to-chum' score periodically to each subscriber of a mobile operator. The results indicate that both decision tree and neural network techniques can deliver accurate chum prediction models by using customer demographics, billing information, contract/service status, call detail records, and service change log. (c) 2005 Elsevier Ltd. All rights reserved.