Sentimen Analisis Berbasis Aspek Berdasarkan Opini Penonton terhadap Stasiun TVRI Menggunakan Adaboost

Abstract

The television industry become a very important sector in the world of entertainment and media. This includes the production, distribution and broadcasting of television programs, as well as the development of the technology that supports them. Nowadays, television has become one of the most popular and effective media in channeling entertainment and informative news to people especially Indonesia. It is become a main idea for the television station is to always find the right and fast solution in responding to the comments and suggestions from the viewers. This is important so that television stations can continue to provide the best service to viewers. To respond to reviews and suggestions from viewers, television stations can take various ways. One of them is by providing a communication facility that allows viewers to provide input, either by telephone, email, or through social media, so there is a need for an approach that is taken by assisting research with the Aspect-Based Sentiment Analysis system using the Adaboost method based on public opinion on TVRI stations. The data will come from public opinion via Twitter and the website directly from TVRI with a total data of 1250 and a comparison of 8:2 between training data and testing data. The average accuracy for the overall sentiment analysis results obtained is 86%.