SPONGY (SPam ONtoloGY): email classification using two-level dynamic ontology.

Area of Science:

• Computer Science
• Information Security

Background:

• Email is a primary communication method, but increasing misuse leads to high volumes of spam.
• Existing spam filtering techniques struggle to keep pace with evolving spam tactics.

Purpose of the Study:

• To develop and evaluate a novel ontology-based multilevel spam filtering system.
• To improve spam filtering accuracy beyond existing methods through a customized approach.

Main Methods:

• Implementation of a two-level ontology spam filter: a global filter and a user-customized filter.
• The global filter utilizes a comprehensive ontology for broad spam identification.
• The user-customized filter adapts to individual user backgrounds and global filter mechanisms.

Main Results:

• The global ontology filter achieved approximately 91% spam filtering, comparable to existing methods.
• The user-customized ontology filter demonstrated significant improvements in spam filtering accuracy.
• The system proved to be scalable and modular, suitable for integration into various platforms.

Conclusions:

• Ontology-based multilevel filtering effectively enhances spam detection accuracy.
• User-customized ontologies provide a powerful tool for personalized and efficient spam management.
• The developed spam filter offers a scalable, modular, and high-performance solution for email security.