• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site
  • HSE University
  • Student Theses
  • Development of a Predictive Model for Determining Churn Willing Clients of the Telecommunication Company Using the XGBoost Algorithm

Student
Title
Supervisor
Faculty
Educational Programme
Final Grade
Year of Graduation
Dmitriy Moshoglo
Development of a Predictive Model for Determining Churn Willing Clients of the Telecommunication Company Using the XGBoost Algorithm
Business Informatics
(Master’s programme)
2018
To date, telecommunications companies share the broadband Internet access market (BBA). The largest part of new connections is generated by attracting customers from other operators with the use of dumping, which means a payback period of up to 1.5 years. The fight against outflow is an alternative approach. PJSC MGTS has implemented a toolkit to identify inclined to outflow subscribers based on the trigger model. It shows insufficient predictive power and low financial efficiency in use. For this reason, there is a need to use new approaches for the company to the implementation of the toolkit based on methods of data mining.

The object of the study is the company PJSC MGTS. The subject of the study is the outflow of subscribers for the BBA. The purpose of the study is to develop a tool for predicting the outflow of subscribers based on a retrospective information provided by them. The development involves a number of tasks: selecting the most appropriate data analysis algorithm for this toolkit, developing a database with a retrospective information on subscribers, preliminary data processing, constructing a predictive model and evaluating its indicators.

The first chapter describes the object, subject, purpose and objectives of the study, reflects the theoretical aspects of potentially applicable data analysis algorithms. In the second chapter, a comparative analysis of algorithms is carried out, the proposed approach to solving research problems and input data for analysis are described. The third chapter is devoted to the practical part of developing the toolkit. The conclusion reflects the main findings of the study.

The resulting parameters of the developed predictive model are satisfactory, the concentration of churn subscribers in the samples is 42%, which is 12% more than the minimum threshold. The toolkit meets all specified requirements. It is planned to introduce the model and integrate it into the current IT landscape of the company.

Student Theses at HSE must be completed in accordance with the University Rules and regulations specified by each educational programme.

Summaries of all theses must be published and made freely available on the HSE website.

The full text of a thesis can be published in open access on the HSE website only if the authoring student (copyright holder) agrees, or, if the thesis was written by a team of students, if all the co-authors (copyright holders) agree. After a thesis is published on the HSE website, it obtains the status of an online publication.

Student theses are objects of copyright and their use is subject to limitations in accordance with the Russian Federation’s law on intellectual property.

In the event that a thesis is quoted or otherwise used, reference to the author’s name and the source of quotation is required.

Search all student theses