Performance comparison classification of spam ham messages with the use of machine learning models

Abstract

The rapid growth in mobile device usage has led to a significant increase in SMS (Short Messaging Service) communication. People now use smartphones daily to send and receive messages, contributing to a surge in unsolicited texts, commonly known as spam. These unwanted messages are often sent by scammers to extract sensitive information such as personal data and financial details, including bank account numbers and credit card information, or for commercial purposes. This has driven the need for effective spam filtering techniques. To address this issue, several machine learning algorithms such as Random Forest, Support Vector Machines (SVM), and Logistic Regression have been applied to distinguish between spam messages and legitimate (ham) ones. The proposed study developed a data driven spam detection system and evaluated the effectiveness of these models. Among those trained, Random Forest and SVM achieved the highest accuracy of approximately 97%. As internet usage grows and organizations increasingly share sensitive data, the threat of SMS spam is expected to rise, making intelligent filtering systems more essential than ever.