User Behavior Based Clusterinf And A Decision Tree Model For Predicting Customer Insolvency In Telecommunication Business

This research project deals with a telecommunication application that had an objective of building a prediction model to predict solvent and insolvent customers in telecommunication business. This focuses on two main data mining techniques, clustering of the customer base to identify the significant characteristics of insolvent customers using an unsupervised model and classification of the customers as solvent and insolvent using a supervised learning method. This classification model can be then be used to predict insolvent customers much earlier than it is done today, so that the company can take preventive measures to reduce the losses. The results of the project show that the model built in this research is a useful tool in the decision making process.