Abstract
Rumors spread among the public that there was fraud occurring during the ongoing flash sale Shopee process. In order to answer the problems found, this study aims to find out how the public feels about the flash sale activity. Data was taken using the Twitter API, and the tweets used started from January 2021–June 2021. As a result, there were 3,467 Twitter data points. After passing through the text preprocessing stage, the data becomes clean data, which amounts to 457. The research results obtained using the Nave Bayes Classifier method show the results of public opinion on Twitter with positive sentiment categories of 19.4%, negative sentiment categories of 43.1%, and neutral sentiment categories of 37.4%, with an accuracy rate of 98%.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
