“…Prinzie and Van den Poel (2005) used constrained optimization to evaluate an individual-level response model for direct marketing, as it focuses on a classification problem applied to direct marketing instead of customer satisfaction, a direct comparison is not possible. Other works such as Xia and Jin (2008), Zhang et al (2012), Vafeiadis et al (2015 and Ismail et al (2015) used techniques such as SVM, ANN, logistic regression, decision trees and rule-based decision process to estimate customer churns. Also, Park and Gates (2009) and Blockhorst et al (2017) applied machine learning techniques to the field of customer relationship to estimate customer satisfaction by analyzing call transcripts using techniques such as decision tree, logistic regression, SVM, ridge regression, lasso regression and random forest.…”