<> "The repository administrator has not yet configured an RDF license."^^ . <> . . . "Sentiment Analysis of the RUU Perampasan Aset Using a\r\nComparative Approach on Bidirectional Encoder\r\nRepresentations from Transformers (BERT) Models"^^ . "Sentiment Analysis of the RUU Perampasan Aset Using a\r\nComparative Approach on Bidirectional Encoder\r\nRepresentations from Transformers (BERT) Models\r\nResi Alindiana1*, Arie Qur'ania2\r\n, Heri Bambang Santoso3\r\n1 Department of Computer Science, Faculty of Mathematics and Natural Sciences, Pakuan University, Bogor, Indonesia\r\n2 Department of Computer Science, Faculty of Mathematics and Natural Sciences, Pakuan University, Bogor, Indonesia\r\n3 Department of Computer Science, Faculty of Mathematics and Natural Sciences, Pakuan University, Bogor, Indonesia\r\n* Correspondence: [065121081@student.unpak.ac.id]\r\nAbstract: The RUU Perampasan Aset has existed since 2003 but has not yet been enacted and continues to attract\r\npublic attention, particularly on the social media platform X (Twitter), alongside the increasing discussion of\r\ncorruption cases. Therefore, this study aims to analyze public perception of the RUU Perampasan Aset from a\r\nlinguistic perspective. Sentiment analysis is conducted by comparing three BERT-based models—mBERT,\r\nIndoBERT, and IndoBERTweet which are capable of understanding bidirectional textual context. Sentiment\r\nclassification is divided into positive and negative categories. Positive sentiment represents supportive opinions\r\nexpressed in polite and ethical language, while negative sentiment represents supportive opinions conveyed in\r\na sarcastic and pessimistic manner. The research process follows the Knowledge Discovery in Databases (KDD)\r\nmethodology. Evaluation results using variations of epochs and batch sizes show that IndoBERT achieves the\r\nbest performance with an accuracy of 92%, followed by mBERT at 91% and IndoBERTweet at 90%. These\r\nfindings indicate performance differences among the three models, suggesting that model effectiveness is\r\ninfluenced not only by architectural design but also by dataset characteristics .\r\nKeywords: RUU Perampasan Aset, Sentiment Analysis, X (Twitter), mBERT, IndoBERT, IndoBERTweet."^^ . "2025-06-16" . . "Universitas Pakuan"^^ . . . "Fakultas Matematika dan Pengetahuan Alam, Universitas Pakuan"^^ . . . . . . . . . . . "Arie"^^ . "Qur'ania"^^ . "Arie Qur'ania"^^ . . "Resi"^^ . "Alindiana"^^ . "Resi Alindiana"^^ . . "Heri Bambang"^^ . "Santoso"^^ . "Heri Bambang Santoso"^^ . . "Universitas Pakuan"^^ . . . "Fakultas Matematika dan Ilmu Pnegetahuan Alam"^^ . . . "Program Studi Ilmu Komputer"^^ . . . . . . "HTML Summary of #10657 \n\nSentiment Analysis of the RUU Perampasan Aset Using a \nComparative Approach on Bidirectional Encoder \nRepresentations from Transformers (BERT) Models\n\n" . "text/html" . . . "Ilmu Komputer"@en . .