SMS Spam Detection with Deep Learning Model

Abstract

The number of mobile users is growing all the time, as well. SMS, which "short messaging service," enables users of both smartphones and conventional phones to send and receive text messages. Because of this, the number of SMS messages experienced a significant increase. In addition to that, the number of unwanted messages known as spam increased. The spammers' purpose is to distribute unsolicited electronic messages for commercial or financial gain, such as market penetration, the purchase of lottery tickets, or the disclosure of credit card information. As a direct consequence of this, sifting through spam receives additional attention. Several different machine learning and deep learning techniques, which are detailed in this Paper, were utilized to detect SMS spam. We developed a spam detection system based on data collected by the University of California, Irvine (UCI). This research study investigates the efficacy of several supervised machine learning algorithms, including the nave Bayes Algorithm, support vector machines, and the maximum entropy algorithm, in detecting spam and ham communications. Additionally, the outcomes of the detection of these messages are displayed here. SMS spam is becoming more prevalent as an increasing number of people use the Internet, and many enterprises share their personal information. SMS spam filtering inherits a substantial amount of functionality from e-mail spam filtering. When evaluating the effectiveness of various supervised learning strategies, the support vector machine method yields the most precise results.