Efficient E-Mail Spam Detection Strategy Using Genetic Decision Tree Processing with NLP Features

Area of Science:

• Computer Science
• Information Security
• Artificial Intelligence

Background:

• Email remains a primary professional communication channel, making it a target for spam and scams.
• While text email spam is researched, voice email security is an emerging challenge.
• Voice emails, or voicemails, deliver speech-based information, presenting unique security vulnerabilities.

Purpose of the Study:

• To develop a robust mechanism for detecting spam in both text and voice emails.
• To address the security gap in voice-based email communication.
• To propose a hybrid data processing approach for comprehensive email spam identification.

Main Methods:

• A hybrid data processing mechanism named Genetic Decision Tree Processing with Natural Language Processing (GDTPNLP) was developed.
• The GDTPNLP approach integrates text and voice data processing for spam detection.
• The system was evaluated for its effectiveness in identifying spam in various email formats.

Main Results:

• The proposed GDTPNLP approach demonstrated a higher spam detection rate.
• Improvements were observed in text extraction speed, overall performance, and cost efficiency.
• The accuracy of spam identification was significantly enhanced by the hybrid method.

Conclusions:

• GDTPNLP offers an effective solution for identifying spam in both text and voice emails.
• The method provides a significant advancement in securing email communication channels.
• The research highlights the importance of hybrid approaches for modern cybersecurity challenges.