Detailed Experimental Results for SEA'17 Publication "Improving Coarsening Schemes for Hypergraph Partitioning by Exploiting Community Structure"

This website accompanies our SEA'17 publication. We provide interactive visualizations of the performance plots as well as detailed per-instance results in .csv format.

Comparison of Edge Weighting Schemes

The performance plots comparing the different edge weighting schemes on the benchmark subset can be viewed here:

• Low Density
• Medium Density
• High Density
• High Density comparing and

Comparison of KaHyPar-CA and other Systems

The performance plots comparing KaHyPar-CA with KaHyPar, hMetis-R, hMetis-K, PaToH-Q, and PaToH-D can be viewed here:

• All Instances
• DAC2012
• ISPD98
• SAT14 Primal
• SAT14 Literal
• SAT14 Dual
• UF-SPM
• Web/Social

Detailed Per-Instance Results

We provide detailed per-instance results for each hypergraph partitioner. For each instance, the .csv files contain the following information:

• hypergraph name
• instance type
• minimum cut (connectivity metric): min_km1
• average cut (connectivity metric): avg_km1
• average imbalance
• average running time

The .csv files for the full benchmark set can be found here:

• KaHyPar-CA
• KaHyPar
• hMetis-R
• hMetis-K
• PaToH-Q
• PaToH-D

Benchmark Instances

The full benchmark set containing all 488 hypergraph can be downloaded here (~3GB). Detailed statistics of the hypergraphs used in the experimental evaluation can be found here in .csv format. A list of hypergraphs used in the benchmark subset can be found here.