The application of data mining for the trouble ticket prediction in telecom operators

Telecommunication Providers face many challenges in retaining customers especially under the fierce competition between each other. One of these challenges is to improve the efficiency of operations and to achieve better customer experience through minimizing the business impact of service interruption and proactively handle it. In this paper, trouble ticket repeat prediction framework is proposed to integrate the enormous data of operations that consist of the voice of the customer (VOC) and the voice of machine (VOM). The VOC is represented by customer trouble tickets collected by the customer relationship management system (CRM) and VOM is represented by the modems' event log that is loaded daily into our platform. Finally, the results from deploying the proposed framework in one of the biggest Telecom Operator in the middle east are presented. In summary, the proposed framework is explained in details starting from tickets classification ending with the actions that control and sustain the improvements through the set up for continuous mining of the data and the plan for monitoring and maintenance of the model.