Consumer Behavior
Author: Sonia Bakhshian (Auburn University)
The purpose of this study was to explore consumers’ criteria and challenges as well as the major defects of face masks at the post-purchase phase. In this study, we also attempted to introduce the application of natural language processing (NLP), as a time- and budget-saving data coding method, to the apparel and textile field. Using secondary data, post-purchase review comments of the top five brands of disposable masks were collected for this study. we utilized the natural language processing (NLP) toolbox and “multinomial nave Bayes” (MNB) machine learning classification models in Python to speed up the coding process and classify consumers' text-based comments. As the result, three models, naming functionality (i.e., comfortability, breathability, protection), defects (i.e., quality, size/fit, ear-loops/nose-wire), and features (i.e., price, country of origin) were identified. The outcome reflected consumers’ needs, expectations, and major current defects in facemasks available in the market. Implications and future research agendas were also discussed.
Keywords: Face-Mask, consumers’ expectations, Natural Language Processing, COVID-19
How to Cite: Bakhshian, S. (2022) “Do Face Masks Really Meet Consumers’ Expectations? Addressing Consumers’ Post-Purchase Concerns and Criteria of Face Masks by Using Natural Language Processing”, International Textile and Apparel Association Annual Conference Proceedings. 78(1). doi: https://doi.org/10.31274/itaa.13575
None