Building a Machine Learning Model to Forecast Consumer Revenge Spending Behavior in the Post COVID-19 Travel Industry

Abstract

This paper aims to build a machine learning model to forecast consumer revenge spending behavior in the post Covid-19 travel industry. Covid-19 has created a new phenomenon "Revenge Spending", where consumers spend excessively in order to compensate for the negative emotion and constraint experienced during the pandemic. This study utilized travel related factors like travel intentions and financial variables to train the machine learning models which included Logistic Regression, Random Forest and Decision Trees. To evaluate which predictive model performs the best in predicting consumer revenge spending behavior in post pandemic travel, cross-validation techniques, accuracy, precision, recall, F1-score, and AUC-ROC metrics were used. The findings of the study bring a meaningful understanding of consumer revenge spending behavior in travel and offer some insight on the key features that are influencing this behavior. Keywords: consumer revenge spending, post Covid-19, travel industry, machine learning, Logistic Regression, Random Forest, Decision Trees, predictive modeling, evaluation metrics, travel behavior