Sentiment Analysis on Twitter Social Media towards Shopee E-Commerce through Support Vector Machine (SVM) Method

Abstract

Shopee is e-commerce widely accessed and used in this era. Many people use Shopee because the products offered are cheaper and more affordable. Despite the fact that Shopee is a well-known e-commerce, it still requires responses and suggestions from the public to maintain or improve the features required. In this study, public sentiment analysis was carried out on Twitter social media related to the Shopee marketplace. This study collected data that contained tweets from predetermined keywords and used Word2Vec and Support Vector Machine classification methods. The use of Word2Vec influenced the level of accuracy so that it increased for each SVM kernel. Meanwhile, the best hyperparameter tuning was found in the polynomial kernel, with an accuracy rate of 93.20%.