Is the Internet in Indonesia Has a Good Sentiment from Netizen?

Abstract

During this pandemic, the Internet has become a basic necessity that equals clothing, food, and shelter. People use the Internet for various activities such as ordering food, communicating, learning remotely (online), working, and social media, one of which is Instagram. On Instagram, people can upload pictures, videos, and even text if they already have an account. This feature is not only limited to an individual but can also be used by companies or, in this case, internet providers. On Instagram comments on internet provider accounts, there are often positive and negative comments that overlap in the comments section, making it difficult for providers to measure the satisfaction of their customers. Therefore in this report, researchers will use sentiment analysis with the Naïve Bayes algorithm and Support Vector Machine (SVM) to measure customer satisfaction through the many hate speeches in the Instagram comments section of internet provider accounts. The research will begin by collecting 1000 data on the internet provider Instagram comments (Text Mining), then the existing data will enter the sentiment analysis process. After that, the sentiment results will be inputted into two algorithms, namely Naïve Bayes and Support Vector Machine (SVM). Then the accuracy will be calculated using the K-Fold Cross Validation and Confusion Matrix to become a precise result. The research results will be in the form of a percentage of customer satisfaction and dissatisfaction based on the many hate speeches that each internet provider has. The results of the analysis of this study are expected to help internet providers in Indonesia know that many customers who are not satisfied with their services can improve the quality of their services.