An intelligent system for churn prediction and customer retention : the case of a telecommunications company | |
| Author | Sarangi, Parth |
| Call Number | AIT Thesis no.IM-18-01 |
| Subject(s) | Machine learning Data mining Neural networks (Computer science) Customer relations--Management |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Information Management, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Thesis ; no. IM-18-01 |
| Abstract | The telecommunications industry is very competitive in most of the developed and developing countries. A few companies operate to provide numerous services to a huge consumer base. Rapid technological advancements in the ICT sector has increased the availability and affordability of mobile telephony devices. With an increasing adoption of mobile telephony devices there is a greater increase in the demand for mobility services. Services such as phone calls, sms, internet etc., have increased over the past decade. With the growing demand and growing consumer base, services providers have reduced prices of these services. Profitability of companies is decided by the number of consumers. In these highly competitive market, customer satisfaction and retention is of high importance. In this study an intelligent churn prediction and customer retention (ICPCR) system is developed. An open source dataset of call detail records (CDR), with 5000 records and 21 features is selected for purpose of system design and model development. Eight data mining methods are employed to generate prediction models. Prediction models such as decision tree, random forest, support vector machines, neural networks and naive bayes are compared for performance and evaluated based on accuracy, sensitivity, specificity and positive prediction metrics. Based on the performance decision tree is selected as the prediction engine in implementation of ICPCR. The customer retention system is designed on decision table rules and upselling techniques are suggested for every customer predicted as churner. This study is designed to facilitate the customer service representative with a mechanism to visualize customer call detail data. In addition, it also enables them to predict the churning status of current customers, and also presents them with an option to control churning by suggesting marketing solutions or product benefits. This study presents a web based approach with KPI dashboards, charts and maps, OALP for roll-up and drill-down analysis, prediction and finally retention strategies for controlling customer churn |
| Year | 2018 |
| Corresponding Series Added Entry | Asian Institute of Technology. Thesis ; no. IM-18-01 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Information Management (IM) |
| Chairperson(s) | Vatcharaporn Esichaikul; |
| Examination Committee(s) | Daily, Matthew N.;Guha, Sumanta; |
| Scholarship Donor(s) | Asian Institute of Technology Fellowship; |
| Degree | Thesis (M. Eng.) - Asian Institute of Technology, 2018 |