Hilbert–Huang Transform Framework-Based Email and SMS Spam Detection

Over the past few decades, there has been an excessive expansion in the use of email and short message service (SMS). While 99% of cell phone users read their SMS messages every day, 90% of people do not read emails. These communication channels, however, are dangerous and can result in spam, which is a kind of hostile attack. Phishing emails are those that purport to be from reputable companies and ask for “financial or personal information.” Some links in these emails could lead to consumers downloading harmful software to their machines. The majority of methods and models created to automatically identify these “SMS and emails” have not yet attained 99.7% accuracy. Lower accuracy has been shown in prior studies when spam detection was done with a limited dataset. In order to address this issue, a classifier of the Hilbert–Huang transform was used in this work to detect spam more accurately on the SMS and email dataset. The researchers came to the conclusion that the suggested approach outperformed earlier models in terms of accuracy and performance after running tests on the real dataset. In particular, the classifier using the Hilbert–Huang transform fared better than the rest. These findings imply that the Hilbert–Huang transform is the best option for categorization.