Analisis Sentimen Pada Sosial Media X terkait Serangan RANSOMWARE Bank Syariah Indonesia (BSI) menggunakan Metode Long Short-Term Memory (LSTM)

This study aims to examine public sentiment regarding the ransomware attack incident on Bank Syariah Indonesia (BSI), by employing the Long Short-Term Memory (LSTM) method combined with Global Vectors for Word Representation (GloVe) for word representation. Data sources are obtained from the social media platform X (formerly Twitter), which underwent preprocessing stages including cleaning, tokenization, stopword removal, and padding. The Random Oversampling (ROS) technique is applied to address the imbalance in sentiment class distribution. Model performance evaluation involves metrics such as accuracy, precision, recall, and F1-score. The test results indicate a significant improvement after ROS implementation, with accuracy reaching 0.94 and F1-score at 0.97, along with the highest recall of 0.98 for the positive class. These findings affirm the superiority of LSTM in detecting sentiment patterns from social media texts, particularly for understanding public reactions to cybersecurity risks. Further analysis reveals the prevalence of negative sentiment post-incident, which may affect public trust in banking stability.