Shopee Customer Sentiment Analysis on Twitter with Naive Bayes Algorithm

Abstract

Sentiment analysis is a field of study that analyzes a person's opinions, evaluates, judges, behaviors, and emotions such as products, services, events, and topics. Sentiment analysis in the business world is usually used to analyze the needs of the community and market needs, which are expected to be able to develop marketing strategies that can increase their company's income. This study takes the data from the Shopee group on Twitter. Shopee is one of the market places that are often used by the people of Indonesia and people in other countries. Shopee's market place that sells services or sells goods for daily needs such as electronics, cutlery, credit, as well as airline tickets, train tickets, and many more services sold by Shopee. The purpose of this research is to find out how many reviews of positive comments and negative comments on the Shopee group on Twitter social media, the data collection uses Rstudio. The Rstudio application can be run on Windows, Linux, or Apple operating systems. As well as for the calculation process using the Naive Bayes method which is included in the sentiment analysis where data collection is taken from the Shopee group. In calculating the level of accuracy using the convusion matrix. The results of this study are that it can be seen that the number of positive and negative comments is equal, namely 150:150, because the data taken is 300 data and the level of accuracy is 97%.