Predictive modelling of user engagement for subscription retention

Thumbnail Image

Files

master_Jayanegara_Angeline_2025.pdf (1.38 MB)

URL

Journal Title

Journal ISSN

Volume Title

School of Science | Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.

Authors

Date

2025-07-31

Department

Major/Subject

Machine Learning, Data Science and Artificial Intelligence

Mcode

Degree programme

Master's Programme in Computer, Communication and Information Sciences

Language

en

Pages

41

Series

Abstract

The rise of digital news and media subscriptions has made subscription based business models increasingly important for news and media organisations such as Sanoma. In this context, understanding and predicting subscription churn of Helsingin Sanomat (main product of Sanoma Media Finland and one of the largest news provider in Finland), as well as identifying the key factors behind it, is essential for ensuring the long term sustainability of the digital media. This thesis investigates subscription retention and churn through predictive modelling, comparing a classical logistic regression model with a more advanced ensemble method, XGBoost. The main objective was not only to improve the accuracy of subscription churn prediction but also to interpret the driving factors behind the churn, thus providing insights for retention strategies. The analysis presents that XGBoost outperformed logistic regression by effectively capturing nonlinear relationships and interactions within the data. For model interpretation, feature importance analysis, gain and SHAP values were applied to the XGBoost model. The SHAP analysis indicated that customers with longer subscription histories and higher engagement activities were less likely to churn. However, these results should be interpreted as insights rather than causal conclusions.

Description

Supervisor

Vehtari, Aki

Thesis advisor

Rautio, Sini

Keywords

machine learning, data science, logistic regression, XGBoost, subscription retention, feature importance analysis

Other note

Citation

Permanent link to this item

https://urn.fi/URN:NBN:fi:aalto-202508196451

Collections

[dipl] Perustieteiden korkeakoulu / SCI