SMS Spam Detection Using Deep Learning Approach

Abstract

AbstractThe popularity of mobile phones has increased drastically in the recent years which is making users vulnerable to various threats like SMS spam, where the user is deceived into revealing private information that could result in a security breach. The motivation of this research is to curb the attackers, hackers, etc., from using SMS spam to exploit mobile device users. Several researchers proposed various machine learning models to automatically detect spam, but they could not achieve a commendable accuracy rate. In this research, several machine learning and deep learning models are utilized to detect SMS spam. A dataset from UCI is used and deep learning models are developed to detect and classify SMS spam using LSTM and BERT. The results are compared with the previous models in SMS spam detection. The proposed deep learning approach obtained the highest accuracy of 99.28% using BERT and 98.84% using LSTM. We utilized Python for all implementations.KeywordsSMS spamDeep learningMachine learningLSTMBERT