An Experimental Analysis of Churn Prediction Techniques on Real Time Datasets
Suraj Saklani1, Shubhangi Neware2
1Suraj Saklani, M. Tech Scholar, Nagpur University, Ramdeobaba college of Engineering & Management, Nagpur, India.
2Dr. Shubhangi Neware, Assistant Professor, Nagpur University, Ramdeobaba college of Engineering & Management, Nagpur, India.
Manuscript received on 08 June 2019 | Revised Manuscript received on 13 June 2019 | Manuscript Published on 08 July 2019 | PP: 252-257 | Volume-8 Issue-8S3 June 2019 | Retrieval Number: H10690688S319/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Churn prediction is an indicative of the loyalty with which the customer is attached to a particular provider. Usually churn or customer churn is a value in percentage, and can be used by various service providers to make sure that the customer stays with them for a longer duration. Based on this value, companies device customer specific plans for higher churning customers, and plans for the customers which are about to opt for another service provider. In this paper, we review and study multiple techniques for customer churn prediction and their application areas, in order to evaluate the techniques and form a basis on which techniques can be used for which particular type of application. Machine learning approaches are generally preferred over traditional ones, as they allow the service providers to learn about the customer behaviour pattern over a long span of customer service usage. We conclude the paper which some suggestions on how churn prediction can be improved for better optimization of the developed system.
Keywords: Churn, prediction, loyalty, machine, learning
Scope of the Article: Machine Learning