Spam Detection in Short Message Service (SMS) Using Naïve Bayes, SVM, LSTM, and CNN
Edward Wijaya, Gracella Noveliora, Kharisma Dwi Utami, Rojali Rojali, Ghinaa Zain Nabiilah · 2023
Short Message Services (SMS) has been most people’s habits. The numbers of SMS usage have been growing rapidly since text messages are more effective than emails. However, there is a problem that people often encounter while using SMS which is spam. Many researchers have discussed what are the best methods to avoid spam, especially in SMS. They attempt to avoid spam by making an SMS filtering machine using various algorithms. In this study, we’re aiming to compare the accuracy percentage of SMS spam detection using various machine learning algorithms, such as Naive Bayes, Support Vector Machine (SVM), Long-Short Term Memory (LSTM), and Convolutional Neural Network (CNN). These algorithms have been implemented and tested over a dataset consisting of 5.574 records. This testing gave specific results of mean, which are 96.95% by using the Naive Bayes Algorithm, followed by 97.93% by using Support Vector Machine, 98.57 by Convolutional Neural Network, and 99.1% by Long-Short Term Memory leads in this testing.