Twitter Sentiment Analysis of Indonesian Airlines Using LSTM

Abstract

Twitter is one of the social media that is currently a trend, where Twitter users can tweet as freely as possible about their opinions and even those opinions about airlines in Indonesia. Twitter sentiment analysis is a process to identify whether tweets on Twitter are included as positive tweets or negative tweets. In this research, the tweets will be divided into three categories: positive, neutral, and negative, using Lexicon and Long Short-Term Memory (LSTM). The data taken are tweets from Twitter in the form of text. One hundred positive, one hundred neutral, and one hundred negative tweets were taken. After going through the process using the Lexicon and LSTM method, the results obtained are 73% accuracy, where there are 130 positive tweets, 105 negative tweets, and 62 neutral tweets.