Machine Learning-Based Detection of Spam Emails

Abstract

Social communication has evolved, with e-mail still being one of the most common communication means, used for both formal and informal ways. With many languages being digitized for the electronic world, the use of English is still abundant. However, various native languages of different regions are emerging gradually. The Urdu language, coming from South Asia, mostly Pakistan, is also getting its pace as a medium for communications used in social media platforms, websites, and emails. With the increased usage of emails, Urdu’s number and variety of spam content also increase. Spam emails are inappropriate and unwanted messages usually sent to breach security. These spam emails include phishing URLs, advertisements, commercial segments, and a large number of indiscriminate recipients. Thus, such content is always a hazard for the user, and many studies have taken place to detect such spam content. However, there is a dire need to detect spam emails, which have content written in Urdu language. The proposed study utilizes the existing machine learning algorithms including Naive Bayes, CNN, SVM, and LSTM to detect and categorize e-mail content. According to our findings, the LSTM model outperforms other models with a highest score of 98.4% accuracy.

A Novel Fuzzy-Logic-Based Multi-Criteria Metric for Performance Evaluation of Spam Email Detection Algorithms

“…Five performance measures are employed, which are accuracy, precision, recall, specificity, and F-score. Siddique et al [12] employ NB, CNN, SVM, and LSTM for spam email detection in Urdu. A self-collected dataset named "Urduemaildataset" is used for evaluation.…”

“…An LSTM network accepts sequential time series data as input. The use of LSTM for email spam detection has been reported in several studies [10][11][12]. A detailed discussion on LSTM can be found in Hochreiter and Schmidhuber [45].…”

“…In particular, artificial intelligence (AI)-based methods have received notable attention from researchers in recent years. Special emphasis has been placed on the methods based on machine learning [1,[10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25]. In addition, deep learning methods have recently been successfully applied to spam email detection [10][11][12][13][14][26][27][28].…”

“…Special emphasis has been placed on the methods based on machine learning [1,[10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25]. In addition, deep learning methods have recently been successfully applied to spam email detection [10][11][12][13][14][26][27][28]. The outcomes of these studies indicate that machine learning and deep learning algorithms provide an effective platform for efficiently solving the spam detection problem; yet there is still a room for further improvement and development.…”

“…The datasets have distinctive features and their suitability and effectiveness is determined by the type of experiments performed to evaluate the performance of the spam detection algorithm. In addition, several researchers use datasets that are not well known, or developed by the researchers themselves [1,12,17,[19][20][21][22]24,28].…”