Machine Learning-Based SMS Spam Detection

AUTHOR(S):

K. Vasumathi, S. Selvakani, K. Santhiya

TITLE

Machine Learning-Based SMS Spam Detection

ABSTRACT

The widespread use of mobile devices has led to a significant increase in SMS (Short Message Service) spam, which undermines the integrity of mobile communication. Unlike internet-based platforms like WhatsApp or Facebook, SMS operates without internet connectivity, making it a unique target for spammers. This study presents a machine learning-based approach to detect and filter SMS spam, addressing the shortcomings of traditional email spam filters which struggle due to limited feature sets, informal language, and a lack of robust SMS spam datasets. The proposed methodology involves the integration of multiple publicly available SMS datasets, followed by data preprocessing, exploratory data analysis, and feature engineering. Various classification algorithms—including Naive Bayes, Support Vector Machine (SVM), and Random Forest—are implemented and compared based on their precision, recall, F1-score, and overall accuracy. Experimental results demonstrate that the proposed models can effectively distinguish spam from legitimate messages, with the SVM model achieving the highest classification performance. These findings have important implications for enhancing mobile security and reducing user exposure to fraudulent or intrusive content.

KEYWORDS

SMS Spam, Facebook, WhatsApp, Internet Connectivity, Financial Gain, Datasets, Data Preprocessing, Feature Engineering, Naive Bayes, Model Development

Cite this paper

K. Vasumathi, S. Selvakani, K. Santhiya. (2025) Machine Learning-Based SMS Spam Detection. International Journal of Computers, 10, 74-80