Sentiment Analysis on the Planned Nickel Mining Development in Raja Ampat Using the Random Forest Algorithm

Abstract

The planned nickel mining development on Kawe and Manuran Islands in Raja Ampat has sparked various public reactions, especially on social media platforms. Raja Ampat is known for having one of the highest levels of marine biodiversity in the world, raising concerns about the potential ecological and social impacts of such development. This study aims to analyze public sentiment regarding the nickel mining plan in Raja Ampat by utilizing social media comments. The method used is the Random Forest algorithm, which is recognized for its high performance in classifying complex text data. A total of 2,010 comments were collected, and after the preprocessing stage, 1,658 clean data entries remained for analysis. The preprocessing steps included text cleaning, case folding, normalization, tokenization, stopword removal, and stemming. The results show that 57.85% of the comments expressed positive sentiment, while 42.15% showed negative sentiment. The Random Forest model was able to classify the sentiments with an accuracy of 80.1%, using three decision trees as the basis for majority voting. Furthermore, n-gram analysis and word cloud visualization provided insight into the dominant words in public opinion, offering a deeper understanding of the issues being discussed. This research is expected to serve as a consideration in development policy-making that prioritizes environmental sustainability and the well-being of local communities.