ConDeNSe Visualizer

Reducing Large Graphs to Small Supergraphs: A Unified Approach

Summarizing a large graph with a much smaller graph is critical for applications such as speeding up graphalgorithms and interactive visualization. In this paper, we propose ConDeNSe (Conditional Diversified Network Summarization), a Minimum Description Length-based method that summarizes a given graph with approximate ‘supergraphs’ conditioned on diverse, predefined structural patterns.

Our focus here is on the 3rd module of ConDeNSe, which selects the structural patterns that will constitute the supernodes of the generated supergraphs. The input is a set of structures, and our methods sift through them and select the ones that will be included in the supergraphs. We have developed three powerful, parallel methods: Step-P, Step-PA and k-Step.

Try the demo!

Select Graph

Nodes:
Edges:

Select Method

VISUALIZATION OUTLINE
  1. The initial graph displays all edges in the input graph in hollow blue circles. These are organized by how the graph has been partitioned by METIS.
  2. At each iteration, the edges of all structures considered by the chosen method are highlighted in green.
  3. The best structure determined by the chosen method has its edges highlighted by a distinct color, demonstrating the spread of the structure across multiple METIS partitions. This process continues until a summarization is reached.