Investigating the Impacts of Customer Experience and Attribute Performances on Overall Ratings Using Online Review Data: Nonlinear Estimation and Visualization with a Neural Network

Abstract

This study discusses interpretable neural networks for marketing and consumer behavior research using customer reviews instead of traditional measurement scales to obtain a better understanding of customer experiences. For comparing the influences of customer experience and service performance on overall satisfaction, the attribute performances are measured by service attribute ratings. Although we use a bag-of-words for the measurements of customer experience, the bag-of-words consists of high-dimensional variables and makes it difficult to identify the effects of each word and service attribute ratings. To solve these problems, this study proposes a useful neural-network method for specifying the expected assumptions based on previous knowledge or theories in consumer behavior research. Because neural networks help estimate nonlinear relationships between objective and predictive variables, a partial dependence plot is applied to visualize the estimated functions and marginal effects. In the empirical analysis, we examine the better model performance and provide reasonable marketing implications from the results of our proposed neural network model.