kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Combining ML Models for Churn Prediction and Customer Retention Strategies within Digital Subscription Services
KTH, School of Electrical Engineering and Computer Science (EECS).
KTH, School of Electrical Engineering and Computer Science (EECS).
2024 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesisAlternative title
Kombinering av maskininlärningsmodeller för churnprognoser och kundbehållningsstrategier inom digitala prenumerationstjänster (Swedish)
Abstract [en]

This bachelor’s thesis investigated the application of machine learning to predict churn and retain customers in digital subscription services. Conducted at KTH Royal Institute of Technology and sponsored by Breakit, a Swedish tech news site, the research developed models with varying feature sets to evaluate performance. These models utilized advanced feature selection methods and machine learning algorithms. The study assessed the feasibility of these models and their potential to recognize and reduce churn, with customer success used as a framework. The results indicate strong performance metrics for the models, indicating significant benefits for the sponsor, including the potential to improve user retention and thus to be able to strengthen their market position in the competitive digital subscription industry.

Abstract [sv]

Denna kandidatuppsats undersökte tillämpningen av maskininlärning för att förutsäga kundavhopp och behålla kunder inom digitala prenumerationstjänster. Studien var genomförd vid KTH Kungliga Tekniska högskolan och sponsrad av Breakit, en svensk nyhetssajt. Modeller med varierande funktionsuppsättningar utvecklades för att utvärdera prestanda. Dessa modeller använde avancerade metoder för urval av funktioner och maskininlärningsalgoritmer. Studien bedömer genomförbarheten av dessa modeller och deras potential att känna igen och minska kundavhopp, med “customer success” som ramverk. Resultaten visar på goda prestationer av modellerna, vilket tyder på betydande fördelar för Breakit, däribland potentialen att förbättra kundbehållning och möjligheten att stärka sin position i den konkurrensutsatta digitala prenumerationsindustrin.

Place, publisher, year, edition, pages
2024. , p. 11
Series
TRITA-EECS-EX ; 2024:249
Keywords [en]
Feature selection, Machine learning, XGBoost, Random Forest, LSTM, Ensemble model, Customer churn, Churn prediction, Customer success, Customer success management (CSM), Customer retention
National Category
Computer and Information Sciences
Identifiers
URN: urn:nbn:se:kth:diva-350656OAI: oai:DiVA.org:kth-350656DiVA, id: diva2:1884564
Supervisors
Examiners
Available from: 2024-08-09 Created: 2024-07-17 Last updated: 2024-08-09Bibliographically approved

Open Access in DiVA

fulltext(627 kB)197 downloads
File information
File name FULLTEXT01.pdfFile size 627 kBChecksum SHA-512
bfa37ccf45dd38bdfeb2447778dd19acca09487f9fe3820a8974cfc01bc6efa5d0859d16905d7b57c06a8540a2f11bdc19ff6d4fbdc37e45c87cdc00f822c024
Type fulltextMimetype application/pdf

By organisation
School of Electrical Engineering and Computer Science (EECS)
Computer and Information Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 197 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 339 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf