Clustering helps to determine the changes in telecom subscribers' behavior

DOI

10.2991/itsmssm-17.2017.70How to use a DOI?

Keywords

clustering, telecom subscribers, churn analysis

Abstract

Telecom company tries to find out new ways for developing new personal-oriented services based on data analysis. Changes in telecom users behavior are the objects of investigation. If a user changes its service usage, the company should pay attention to the fact and minimize the negative outcomes (e.g. churn analysis). We propose a method, which allows identifying changes in user behavior based on analysis of Call Detail Records data. The main idea of the method is using clustering techniques to determine the clusters with typical user behavior. Since the clusters exist and all users are labeled in terms of cluster belonging, the new data about user behavior compare with a typical profile. Open data was used in use cases. The results show the outperform of the proposed method in comparison with benchmark model.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Alexey Golubev
AU  - Ngyuen Anh Tuan
AU  - Maxim Shcherbakov
AU  - Tran Van Phu
PY  - 2017/12
DA  - 2017/12
TI  - Clustering helps to determine the changes in telecom subscribers' behavior
BT  - Proceedings of the IV International research conference "Information technologies in Science, Management, Social sphere and Medicine" (ITSMSSM 2017)
PB  - Atlantis Press
SP  - 339
EP  - 343
SN  - 2352-538X
UR  - https://doi.org/10.2991/itsmssm-17.2017.70
DO  - 10.2991/itsmssm-17.2017.70
ID  - Golubev2017/12
ER  -