E-mail Fraud Detection Using Deep Learning

The issue of spam on social media and applications has gotten worse with the growth of the Internet of Things. To address the issue, researchers have suggested a number of spam detection techniques. One of the reasons spam rates are still high is that most malicious e-mails still link to risky websites, despite the existence of anti-spam tools and tactics. Spam can cause servers to lag by consuming memory or storage. One important method for identifying and eliminating spam is e-mail filtering. We evaluated several machine learning and deep learning methods, including Naive Bayes, decision trees, random forests, and support vector machines, in order to accomplish this goal. Machine learning is heavily used by e-mail filters and Internet of Things spam filters, which fall under this category as well. Studies show that spam is still a major issue on social media sites and applications. Furthermore, spam has proliferated globally due to the increased use of mobile devices and the expansion of e-mail services. This study suggests that the best way to address this issue is to employ various forms of machine learning to identify and eliminate spam. According to experimental findings, TF-IDF random forest classification performs more accurately than alternative algorithms. Accuracy is the only metric that can be used to evaluate performance because the dataset is unequal. As a result, the algorithm needs to be accurate, have a strong recovery, and an F-measure.