A Study on User Perception of Delivery Apps Using Text Mining Techniques : Focusing on Coupang Eats App Reviews

A Study on User Perception of Delivery Apps Using Text Mining Techniques : Focusing on Coupang Eats App Reviews

The purpose of this study is to identify users' perceptions of the Coupang Eats app, which is maintaining growth in the delivery app market, and to derive improvements by identifying factors that induce and strengthen positive/negative perceptions. As a result of the study, first, the frequency and connection centrality of the word 'fast' were high, indicating that the speed of the Coupang Eats service is receiving a lot of attention. Second, as a result of CONCOR analysis, the areas of interest of users were identified by clustering the main keywords of the Coupang Eats app review data and dividing them into groups. A total of four groups could be identified: 'benefit service', 'satisfaction', 'inconvenience', and 'other', indicating the areas of interest of users. Third, through emotional vocabulary analysis, it was confirmed that there were more positive vocabulary than negative vocabulary in the Coupang Eats app review data. Fourth, the result of document sensitivity classification also confirmed that positive reviews appeared more than negative reviews. Finally, a text analysis was conducted on positive and negative reviews classified according to the document sensitivity classification to derive factors that induce and strengthen positive and negative perceptions.