Sentiment analysis data by ShopeePayLater Twitter’s opinion using naïve Bayes classifier

Abstract

The Covid-19 pandemic effect on online shopping transactions rather than offline in order to prevent the spread of the Covid-19 virus. The paylater service help Indonesian become more easier to shop by taking out a loan. Shopee is one of the marketplaces that has a paylater service, namely Shopee PayLater, which is one of the payment methods on the Shopee platform that allows Shopee users to shop and only pay at a later date when it is due. Every new ShopeePayLater user want to know how the previous user responded as a form of testimony after using this feature via Twitter in the form of tweets. So it is necessary to conduct a sentiment analysis to find out public opinion with the existence of ShopeepayLater using the Naïve Bayes Classifier method so that it can provide useful information for Shopee and the public regarding ShopeePayLater services. Calculation of classification accuracy is carried out using G-Mean and AUC, because the sentiment of the tweets data is included in the imbalancedata category resulting in a G-Mean value of 64,79% and AUC of 66,12%.