Performance Comparison of Anti-Spam Technology Using Confusion Matrix Classification

F. Rahmad, Y. Suryanto, K. Ramli

Research output: Contribution to journalConference articlepeer-review

10 Citations (Scopus)

Abstract

This research aims to test the ability to capture malicious e-mail objects, spam e-mails, phishing e-mails, or malware e-mails on Proofpoint Email Protection and Fortimail antispam devices to prevent security threats to PT. XYZ. Email service was one of the primary services for an agency or company in providing electronic mail that could be used to exchange official information. The study was conducted by comparing two antispam devices of Proofpoint and Fortimail. The test was conducted sequentially within three months period on each device. Those devices were configured using the infrastructure of PT. XYZ. Besides comparing the functions or features of each device, device testing was also done on overcoming the problem of email security threats such as spoofing, impostor, and bulk email that often threaten email infrastructure. To measure the accuracy of the results of the two devices, we use the Confusion Matrix method to measure the performance of the classification system. As the result, the slights differences inaccuracy were the indication that the ability of each antispam device was not too far. A large number of spam emails which caught through antispam devices were illustrated that PT. XYZ had become the target of spam, phishing, and malware attack. For this reason, anti-spam device technology was needed to maintain the quality of PT. XYZ email service.

Original languageEnglish
Article number012076
JournalIOP Conference Series: Materials Science and Engineering
Volume879
Issue number1
DOIs
Publication statusPublished - 5 Aug 2020
Event3rd International Conference on Informatics, Engineering, Science, and Technology, INCITEST 2020 - Bandung, Virtual, Indonesia
Duration: 11 Jun 2020 → …

Fingerprint

Dive into the research topics of 'Performance Comparison of Anti-Spam Technology Using Confusion Matrix Classification'. Together they form a unique fingerprint.

Cite this