Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/419430
Title: | Mining Mobile Data for Telecom Churn Management |
Researcher: | Kapoor Vani |
Guide(s): | Mamta Madan and Dave Meenu |
Keywords: | Computer Science Computer Science Hardware and Architecture Engineering and Technology |
University: | Jagannath University, Jaipur |
Completed Date: | 2022 |
Abstract: | Churn is inevitable part of each and every business. It cannot be completely removed. But there is definitely the possibility of reducing it, for the sake of company s profitability. None of the businesses will wish to lose the customers to the competitors. Every business wants to ensure that they retain their customers. So, how to deal with churn by providing different varieties of solutions to it, is referred to, what is known as Churn Management. It may involve, strategy making on how to be ahead of the competitors, that may involve coming up with latest technologies and new and lucrative schemes. newlineFor any service-oriented organization, customers are like assets. Every company may it be small, medium or large scale, survives by winning the hearts of its customers. Losing them to competitors will lead to a financial setback. newlineIf the organization is able to predict the behavior of its customers in advance, then, this loss can be minimized or reduced to a negligible level. This prediction requires the detailed study and analysis of the current customers, and also of the churned customers. This process of identifying the customers who are unhappy and are about to quit, is called Churn Prediction. newlineCurrent research work is carried out in the direction of detecting and predicting well in time, the customers, who are about to churn a particular mobile phone subscriber. This work has shown the way, by which the customers, who are having the probability to churn, can be predicted beforehand by analyzing the behavior patterns of old customers who turned out to be churners in the past. So, by applying the machine learning technique, the system is trained to learn itself from the old behavior, and improve its results over the time. As a technique of machine learning, Decision Tree has been applied for analyzing the past records and finding out the patterns. newlineIn the second phase of research work, genetic algorithm has been used with hill climbing as the technique for optimizing the results of machine learning. Genetic algorith |
Pagination: | |
URI: | http://hdl.handle.net/10603/419430 |
Appears in Departments: | Faculty of Engineering and Technology |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 30.22 kB | Adobe PDF | View/Open |
02_prelim pages.pdf | 551.33 kB | Adobe PDF | View/Open | |
03_content.pdf | 220.95 kB | Adobe PDF | View/Open | |
04_abstract.pdf | 63.76 kB | Adobe PDF | View/Open | |
05_chapter01.pdf | 255.42 kB | Adobe PDF | View/Open | |
06_chapter02.pdf | 339.96 kB | Adobe PDF | View/Open | |
07_chapter03.pdf | 267.39 kB | Adobe PDF | View/Open | |
08_chapter04.pdf | 332.15 kB | Adobe PDF | View/Open | |
09_chapter05.pdf | 168.66 kB | Adobe PDF | View/Open | |
09_chapter06.pdf | 261.58 kB | Adobe PDF | View/Open | |
10_annesxures.pdf | 2.73 MB | Adobe PDF | View/Open | |
10_chapter07.pdf | 224.42 kB | Adobe PDF | View/Open | |
11_chapter08.pdf | 326.97 kB | Adobe PDF | View/Open | |
12_chapter09.pdf | 254.43 kB | Adobe PDF | View/Open | |
13_chapter10.pdf | 129.55 kB | Adobe PDF | View/Open | |
14_chapter11.pdf | 516.78 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 221.52 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: