Compares botnet detection methods by computing the error metrics by reading the labels on a NetFlow file. The original NetFlow should have a new column for the ground-truth label, and a new column with the prediction label for each botnet detection method.

This program computes all the error metrics (TPR, TNR, FPR, FNR, Precision, Accuracy, ErrorRate, FMeasure1, FMeasure2, FMeasure0.5) and output the comparison results. It also ouputs a png plot.

The program can compare in a flow-by-flow basis, or it can apply our new botnet detection error metrics, that is time-based, detects IP addresses instead of flows and it is weighted to favor sooner detections. See the paper for more details.

Project Samples

Project Activity

See All Activity >

License

GNU General Public License version 3.0 (GPLv3)

Follow Botnet Detectors Comparer

Botnet Detectors Comparer Web Site

Other Useful Business Software
Turn Your Content into Interactive Magic - For Free Icon
Turn Your Content into Interactive Magic - For Free

From Canva to Slides, Desmos to YouTube, Lumio works with the tech tools you are already using.

Transform anything you share into an engaging digital experience - for free. Instantly convert your PDFs, slides, and files into dynamic, interactive sessions with built-in collaboration tools, activities, and real-time assessment. From teaching to training to team building, make every presentation unforgettable. Used by millions for education, business, and professional development.
Start Free Forever
Rate This Project
Login To Rate This Project

User Reviews

Be the first to post a review of Botnet Detectors Comparer!

Additional Project Details

Operating Systems

BSD, Linux

Intended Audience

Science/Research, Security, Security Professionals

Programming Language

Python

Related Categories

Python Anti-Malware Software, Python Machine Learning Software, Python Test and Measurement Software

Registered

2014-02-20