<![CDATA[This study aims to analyze and compare the performance of three widely used machine learning algorithms for data classification: Support Vector Machine (SVM), Decision Tree, and Naïve Bayes. These algorithms employ distinct approaches in handling data, making it essential to evaluate their effectiveness and efficiency in classification tasks. In the digital era characterized by massive data growth, the selection of an appropriate classification algorithm is a critical determinant for accurate and efficient data-driven decision-making. The main contribution of this research is to provide a comprehensive understanding of the relative strengths and limitations of each algorithm under varying data conditions. This study not only highlights comparative performance outcomes but also emphasizes practical implications for researchers and data science practitioners in selecting algorithms suited to specific needs. In doing so, it addresses a research gap concerning integrated evaluations of data characteristics and algorithmic performance. The methodology adopts a quantitative approach through computational experiments using standardized datasets (Titanic, Spam Email, and Wine). The datasets were divided into training and testing sets and analyzed using Python with the scikit-learn library. Performance evaluation was conducted based on accuracy, precision, recall, and F1-score, validated through cross-validation techniques to ensure reliability of results. The findings indicate that SVM outperforms in terms of accuracy and recall on complex datasets, Naïve Bayes is more efficient in computational time particularly for text data, while Decision Tree stands out for model interpretability despite slightly lower accuracy. These results are expected to serve as a practical reference for selecting suitable algorithms according to data characteristics, thereby supporting more targeted and intelligent modeling strategies in the era of digital transformation.]]>
