Klasifikasi Sentimen Terhadap Kemunculan Mobil Listrik Menggunakan Metode Naïve Bayes

Abstract

The emergence of electric vehicles has triggered diverse public opinions on social media, reflecting hopes for environmental sustainability alongside concerns regarding functional aspects. This study aims to classify public sentiment toward electric vehicles using the Naïve Bayes algorithm. Data were obtained from the Kaggle platform and processed through pre-processing stages, including cleaning, case folding, stopword removal, and label correction to ensure dataset quality. The analysis was conducted using RapidMiner software by applying the Naive Bayes, Apply Model, and Performance operators. The results indicate that the model achieved a perfect accuracy rate of 100%, with precision, recall, and f1-score values of 1.00 across all classes (positive, negative, and neutral). This achievement is influenced by the characteristics of the dataset, which possesses a clearly separated distribution of linguistic features and minimal noise following the pre-processing stage. Although the model demonstrates exceptional performance, the implementation of k-Fold Cross-Validation techniques is highly recommended to ensure the model's validity and generalizability to more complex real-time data. This study confirms that Naïve Bayes remains a highly competitive instrument for mapping public opinion to support energy transition policymaking