Churn Classification: An Exploration of Features to Improve the Performance

Pedro Shiguihara, Javier Dioses

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This work explores the domain expert's knowledge-based feature engineering for the churn problem. We employ 10-fold cross-validation for parameter tunning and leave-one-out validtion on baselines classifiers. An improvement of up to 9.2% was achieved in terms of the true positive average rate compared to the original dataset, using baselines classifiers. We consider the true positive rate one of the main indicators to measure the churn problem jointly with accuracy and F1. The proposed criteria outperformed the original features between 1.29% and 4.63% in average accuracy using leave-one-out cross-validation. It was also observed that decision trees and ensembles methods performed better than stand-alone classifiers in this problem.

Original languageEnglish
Title of host publicationProceedings of the 2022 IEEE Engineering International Research Conference, EIRCON 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665450829
DOIs
StatePublished - 2022
Event2022 IEEE Engineering International Research Conference, EIRCON 2022 - Lima, Peru
Duration: 26 Oct 202228 Oct 2022

Publication series

NameProceedings of the 2022 IEEE Engineering International Research Conference, EIRCON 2022

Conference

Conference2022 IEEE Engineering International Research Conference, EIRCON 2022
Country/TerritoryPeru
CityLima
Period26/10/2228/10/22

Keywords

  • churn prediction
  • classification
  • feature engineering
  • supervised machine learning

Fingerprint

Dive into the research topics of 'Churn Classification: An Exploration of Features to Improve the Performance'. Together they form a unique fingerprint.

Cite this