Sentiment analysis for ranking online shopping sites in Indonesia using the Naive Bayes method (secondary data case study).

Abstract

Abstract: Lazada, Bukalapak and Blibli.com are companies involved in online markets and malls. To respond to opinions, criticisms and complaints, Lazada, Bukalapak and Blibli.com have a website platform that reviews the services and services offered by the three companies. The website platform is named priceprice.com. The type of comment is a large amount of unstructured text. This condition can cause online shopping companies to miss useful information from a collection of text documents.The setimen analysis is specified in this case study to rank the popularity of online shopping sites in Indonesia. In this study only discussed three online shops, namely Lazada, Bukalapak and Blibli.com.This case study uses the Naïve Bayes Classifier (NBC) method with k-fold cross validation testing to ensure data accuracy. The Naive Bayes method is used because it is more suitable in the operation of sentiment analysis in this study compared to other methods. In this study Lazada is the most popular online shop compared to Bukalapak and Blibli.com. Based on calculation accuracy naïve bayes uses k-fold cross validation for Lazada to have an average 96,5%, Bukalapak 93,9% and Blibli 89,7%.

Keywords: Lazada, Bukalapak, Blibli.com, Naive Bayes