Skip to main content
ClusterHawk
Sign Up

Platform Documentation

Learn how to use ClusterHawk for IP clustering and threat detection

Search Documentation

1
Submit IPs

Upload your IP addresses of interest through our secure interface. Our platform handles datasets up to 5000 addresses.

2
Analysis

Our deterministic ensemble pipeline analyzes patterns, identifies relationships, and generates threat intelligence automatically — same input, same clusters, same reasoning, every run.

3
Receive reports

Get comprehensive threat intelligence reports with IOCs, YARA rules, and hunting queries.

4
Execute hunting queries

Use our automated hunting query execution service to validate findings and monitor for new threats.

User Guide

3D Cluster Visualization

3D Cluster Visualization

The 3D Cluster Visualization gives an interactive three-dimensional view of your clustering results. Use it to explore cluster relationships, patterns, and anomalies visually.

Visualization Modes

Two visualization modes are available:

  • Point Cloud with Convex Hulls: Displays individual IP addresses as colored 3D points, with semi-transparent convex hull meshes wrapping around each cluster to show cluster boundaries. Adjust hull opacity or enable wireframe mode for different views.
  • Cluster Spheres with Orbital IPs: Shows clusters as semi-transparent spheres centered on cluster centroids, with IPs positioned at their actual 3D-reduced coordinates. Includes outlier highlighting for IPs that are spatially outside their cluster sphere boundary.
Dimensionality Reduction Methods

Three dimensionality reduction methods project your high-dimensional clustering data into 3D space:

  • PCA (Principal Component Analysis): Fast linear projection that preserves global structure. Best for quick exploration and overview.
  • t-SNE: Non-linear method that preserves local neighborhood relationships. Good for seeing how individual clusters are structured.
  • UMAP: Balances both local and global structure preservation. Generally provides the most informative view of cluster relationships.

Note: Not all reduction methods may be available for every dataset. Method availability depends on the characteristics of the input data processed by the pipeline.

Interactive Controls
  • Rotate: Click and drag to rotate the view
  • Zoom: Use scroll wheel or the +/- buttons
  • Pan: Right-click and drag to move the camera
  • Center on cluster: Double-click any point, hull, or sphere to focus the camera on it
  • Reset view: Click the reset button to return to the default camera position
  • Auto-rotate: Enable automatic rotation for presentation or exploration
  • Point size: Adjust the size of IP points for better visibility
  • Show/hide noise: Toggle visibility of noise (cluster -1)
Cluster Highlighting

Click on a sphere, hull, or legend chip to highlight a cluster and its contained points. In hierarchical mode, clicking a large sphere highlights all smaller clusters contained within it. Click again to clear the highlighting.

Understanding the Visualization

Keep in mind that this visualization projects high-dimensional feature space into 3D. Due to dimensionality reduction, clusters that are well-separated in the original space may appear closer or overlapping in the 3D view. The spatial relationships shown are approximations. Use this tool for exploration and pattern discovery, but refer to the clustering metrics and feature analysis for definitive cluster quality assessment.