Sentiment Analysis of 'Kampus Merdeka' on Twitter Using Support Vector Machine (SVM) Algorithm

This study aims to analyze public sentiment toward the Kampus Merdeka policy using Twitter data from January 2020 to February 2023. The policy, introduced by the Ministry of Education, Culture, Research, and Technology in Indonesia, allows students to gain real-world experience aligned with their interests as part of career preparation. Sentiment analysis used the keyword "Kampus Merdeka" to classify opinions as positive, negative, or neutral. The CRISP-DM (Cross Industry Standard Process for Data Mining) framework encompasses critical phases, including business understanding, data comprehension, data preparation, modeling, and evaluation. Sentiment labeling was performed using a lexicon-based approach, and the Support Vector Machine (SVM) algorithm was applied for classification. The model's accuracy across different training and test data splits was 89.98%, 90.81%, and 91.18%, with the 90:10 split yielding the highest accuracy. Results showed that 57.17% of the tweets had positive sentiment, 26.77% were negative, and 16.10% were neutral. This study highlights the effectiveness of SVM in sentiment classification.