Analisis Perspektif Mengenai Kebijakan Pemerintah Bergabung Dengan Board Of Peace Menggunakan IndoBERT dan NusaBERT

Abstract

The Board of Peace is an international body initiated by U.S. President Donald Trump. The government’s decision to join this body has naturally sparked debate among the public, particularly on the social media platform X.  The social media platform X (Twitter) is one of the best platforms for monitoring public opinion on various topics, including government policies—such as those implemented by President Prabowo Subianto. Twitter provides users with a space to follow the latest developments, participate in conversations, and express their opinions on public policy and leadership. In Indonesia, Twitter is frequently used for planning social events, sharing information, and voicing criticism or expectations toward national leaders. This study aims to analyze public sentiment regarding Indonesia’s joining the Board of Peace using the IndoBERT and NusaBERT algorithms. The dataset used, consisting of 290 data points, was obtained through data crawling techniques from the social media platform X. The research stages included Data Collection, Data Pre-Processing, Data Balancing, Fine-Tuning of the Transformer Model, and Model Evaluation (accuracy, precision, recall, and F1-score). The results of this study show that the IndoBERT algorithm achieved an accuracy of 90%, while NusaBERT achieved an accuracy of 97%. This indicates that the NusaBERT algorithm has a 7% advantage in accuracy. This study underscores the importance of selecting the appropriate algorithm to enhance the performance of sentiment analysis systems on public policy topics.

GS Cited Analysis