Sentiment Analysis of YouTube Video Comments with the Topic of Starlink Mission Using Long Short Term Memory

The launch of the Starlink Satellite by a private company from the United States (SpaceX) has become a hot topic for conversation. SpaceX is a project that aims to provide satellite-based internet services worldwide with high performance and affordable prices even in remote places. The satellite launch video was shared on the YouTube platform. Many comments were on the satellite launch video, and various reactions were written on the video’s comments column. Therefore, in this study, sentiment analysis will be carried out to analyze the responses of internet users through comments on YouTube to the launch of the Starlink satellite using deep learning. The YouTube comment data used in this study amounted to 22,000 comments. The model used is Long Short Term Memory (LSTM). This study will produce an accuracy value of the LSTM model by applying different activation and optimization functions. This study shows that the highest accuracy is 86% using the LSTM model, Softmax activation function, and Adam’s optimization.