Sentiment Analysis of Public Perceptions toward Finance Minister Purbaya

This dataset contains Indonesian-language social media posts related to Finance Minister Purbaya Yudhi Sadewa collected from the Platform X (formerly Twitter). The dataset was developed for sentiment analysis research using the IndoBERT model. A total of 3,121 posts were collected, with 1,000 manually labeled instances categorized into positive, neutral, and negative sentiments. The posts were obtained through a web crawling process using keyword-based queries to capture discussions surrounding economic and policy-related topics. The preprocessing pipeline includes text cleaning, normalization, and the removal of duplicates, noise, and non-informative tokens. This dataset is suitable for Indonesian Natural Language Processing (NLP) research, particularly for sentiment classification, public opinion analysis, and language model fine-tuning tasks involving social media data.