<![CDATA[The Indonesian Investment Management Agency Danantara program is an initiativeby the Indonesian government aimed at strengthening the national economy through the management of strategic assets and the enhancement of global investment. The launch of this program has generated various public responses on social media, particularly on the X (Twitter) platform. This study aims to analyze public sentiment in Indoensia roward the Danantara program and to compare the performance of two machine learning algorithms, namely Naïve Bayes and Support Vector Machine (SVM). The analysis process begins with data collection through web crawling, followed by data preprocessing, sentiment labeling using a hybrid lexicon-based approach, TF-IDF weighting, and classsifcation using both algorithms. Model Evaluation is conducted using accuracy, precision, recall, and F1-score metrics. The results indicate that the SVM algorithm achieves an accuracy of 87% while the Naïve Bayes algorithm attains an accuracy of 83,7%. These findings demonstrate that SVM outperforms Naïve Bayes in sentiment text classification. The result of this study is expected to serve as a reference for understanding public perception and as a consideration for policymakers.Keywords: Sentiment Analysis; Danantara; Support Vector Machine; Naïve Bayes; Lexicon-BasedAbstrakProgram Badan Pengelola Investasi Danantara merupakan inisiatif pemerintah indonesia yang bertujuan untuk memperkuat perekonomian nasional melalui pengelolaan aset strategis dan peningkatan investasi global. Peluncuran program ini memunculkan berbagai tanggapan di media sosial, terutama di platform X (Twitter). Penelitian ini bertujuan untuk menganalisis sentimen masyarakat Indonesia terhadap program Danantara serta membandingkan kinerja dua algoritma machine learning, yaitu Naïve Bayes dan Support Vector Machine (SVM). Proses analisis dimulai dengan pengumpulan data dengan crawling, preprocessing data, pelabelan menggunakan pendekatan hybrid Lexicon-based, pembobotan TF-IDF, serta klasifikasi menggunakan kedua algoritma. Evaluasi dilakukan dengan metrik accuracy, precision, recall, dan f1-score. Hasil penelitian menunjukkan bahwa algoritma SVM memperoleh nilai akurasi sebesar 87% dan Naïve Bayes memperoleh akurasi sebesar 83,7%. Temuan ini menunjukkan bahwa SVM lebih unggul dalam mengklasifikasikan teks sentimen. Hasil penelitian ini diharapkan dapat menjadi referensi dalam memahami persepsi publik serta menjadi bahan pertimbangan bagi pemangku kebijakan. ]]>
Copyrights © 2025
