Analisis Sentimen Pada Twitter terhadap Isu Penundaan Pemilu 2024 Menggunakan Metode Long Short-Term Memory (LSTM)

Social media is a form of social activity that utilizes online networks to communicate through text, images, and videos. One of the social media platforms that remains a space for individual expression is Twitter. This study examines public sentiment on Twitter regarding the postponement of the 2024 elections in Indonesia. The primary focus of this research is to analyze public opinion on this issue using the Long Short-Term Memory (LSTM) method in sentiment analysis. The data used is derived from tweets related to the postponement of the elections on Twitter. The aim of this study is to understand the extent of public sentiment towards the postponement of the 2024 elections, whether it is positive, negative, or neutral. The method employed is LSTM, a deep learning technique used to process text data in sequence. The data used includes 1758 samples after preprocessing. The research findings indicate that 58.0% of the samples exhibit positive sentiment, while 42.0% show negative sentiment. The LSTM model applied achieves an accuracy rate of 88% in classifying public sentiment on Twitter regarding the postponement of the elections. The conclusion of this study is that the majority of individuals engaged in conversations on Twitter regarding the postponement of the 2024 elections display positive sentiment. However, there is still a portion that expresses negative sentiment. The high accuracy rate of the LSTM model demonstrates the effectiveness of this method in analyzing text data from social media. Therefore, this research provides valuable insights into public perspectives on this important issue in the digital era.