Profit-based churn prediction based on Minimax Probability Machines

dc.coverageDOI: 10.1016/j.ejor.2019.12.007
dc.creatorMaldonado, Sebastián
dc.creatorLópez, Julio
dc.creatorVairetti, Carla
dc.date2020
dc.date.accessioned05-01-2026 18:19
dc.date.available05-01-2026 18:19
dc.description<p>In this paper, we propose three novel profit-driven strategies for churn prediction. Our proposals extend the ideas of the Minimax Probability Machine, a robust optimization approach for binary classification that maximizes sensitivity and specificity using a probabilistic setting. We adapt this method and other variants to maximize the profit of a retention campaign in the objective function, unlike most profit-based strategies that use profit metrics to choose between classifiers, and/or to define the optimal classification threshold given a probabilistic output. A first approach is developed as a learning machine that does not include a regularization term, and subsequently extended by including the LASSO and Tikhonov regularizers. Experiments on well-known churn prediction datasets show that our proposal leads to the largest profit in comparison with other binary classification techniques.</p>eng
dc.identifierhttps://investigadores.uandes.cl/en/publications/b64485fd-02c1-48df-9f71-6be5fdd3a405
dc.languageeng
dc.rightsinfo:eu-repo/semantics/openAccess
dc.sourcevol.284 (2020) date: 2020-07-01 nr.1 p.273-284
dc.subjectAnalytics
dc.subjectChurn prediction
dc.subjectMinimax probability machine
dc.subjectRobust optimization
dc.subjectSupport vector machines
dc.titleProfit-based churn prediction based on Minimax Probability Machineseng
dc.typeArticleeng
dc.typeArtículospa
Files
Collections