The rapid digitalization of Small and Medium Enterprises (SMEs) has led to significant shifts in business operations,
especially in their adaptation to digital platforms. Public perception towards this digital transformation is crucial to understand, as
it reflects the success and acceptance of these efforts. This research conducts sentiment analysis on social media platform X to
classify public opinions regarding the digitalization of SMEs. The analysis employs two machine learning algorithms, Support
Vector Machine (SVM) and K-Nearest Neighbor (KNN), using Term Frequency-Inverse Document Frequency (TF-IDF) for
feature extraction. The study compares the performance of both models under baseline and hyperparameter-tuned conditions. The
results show that the SVM model consistently outperforms KNN in terms of accuracy, precision, recall, and F1-score. The highest
accuracy achieved by the SVM model is 81.97% after hyperparameter tuning with a sigmoid kernel. Meanwhile, the best KNN
model records an accuracy of 81.31% using Manhattan distance with 11 neighbors. This study demonstrates that SVM provides
better stability and performance in sentiment classification related to SME digitalization. The findings are expected to help
policymakers better understand public sentiment and formulate more effective strategies for supporting SME digital
transformation.
