Mining the network value of customers
Proceedings of the seventh ACM SIGKDD international conference on Knowledge discovery and data mining
Customer churn prediction using improved balanced random forests
Expert Systems with Applications: An International Journal
Hi-index | 0.01 |
Nowadays, customer churn is one of the toughest problems. Thanks for the development of database and data mining, the customer churn rate can be predicted exactly. But is the higher churn rate customer should be the most urgent to retain? The customer intention tactics is becoming the core problem for the customer churn management. This paper aimed to introduce the network value to the customer intention tactic of the SCMDA. First, the call information of SCDMA in a city branch of China Telecom was collected, then the customer churn rate is predicted by a Decision Tree Model via SAS8.50, and finally, intention tactic is made based on the network value of complex network and customer churn rate.