<![CDATA[This study discusses the application of the Naive Bayes algorithm to detect spam messages in Short Message Service (SMS) services. The background of this study is the increasing spread of spam messages containing advertisements, fraud, and malicious content, which necessitates an automated system to distinguish spam from non-spam. The methods used in this study include collecting labeled SMS data, preprocessing (text cleaning, tokenization, stopword removal, and stemming), and feature extraction using the Term Frequency-Inverse Document Frequency (TF-IDF) technique. The Naive Bayes model was trained on a Kaggle dataset and tested in Google Colab to evaluate classification performance using accuracy, precision, and recall metrics. The results showed that the Multinomial Naive Bayes model achieved an accuracy of 96.86%, with a strong ability to recognize ham (non-spam) messages and exemplary performance in detecting spam messages. These findings demonstrate that the Naive Bayes algorithm is effective and efficient at classifying Indonesian-language text messages, making it a suitable basis for developing a more innovative, faster automatic SMS spam detection system.]]>
Copyrights © 2025
