Security analysts spend hours manually checking whether suspicious internet addresses are dangerous. This system automates the check by pulling threat ratings from multiple sources and making a quick, reliable call.
The problem
Detecting and classifying malicious IP addresses is essential for enterprise security, but manual reputation checks are slow and inadequate for busy security operations centres.
What the research did
We built an automated, multi-source system that takes user-submitted IP addresses, queries the AbuseIPDB and VirusTotal threat intelligence services, computes a weighted score and classifies each address as “good” or “bad”.
What we found
Experiments showed correct classification and faster IP reputation checks compared with traditional static methods. Analysts can then focus on critical tasks and respond to incidents faster.
Who benefits
- Security operations centres
- IT teams in small and medium businesses
- Incident responders
Limits and what comes next
The system relies on the coverage and accuracy of external services. This conference paper laid groundwork for our more advanced IP SafeGuard framework.
The team
This research was led by Abdullah Al Siam with Md. Maruf Hassan (Southeast University), me, Injamamul Haque Sonet, Abdul Kadar Muhammad Masum, Touhid Bhuiyan and Dewan Md. Farid.
Frequently asked questions
What is IP reputation?
A rating of how likely an internet address is to be involved in malicious activity.
What sources does the system use?
AbuseIPDB and VirusTotal.
How is this related to IP SafeGuard?
IP SafeGuard extends the idea with machine learning and more sources.