Preserving Minority Structures in Graph Sampling

Sampling is a widely used graph reduction technique to accelerate computations and simplify visualizations. By comprehensively analyzing the literature on sampling, we assume that existing algorithms cannot effectively preserve minority structures are rare small in but very important analysis. In this work, initially conduct pilot user study investigate representative most appealing human viewers. We then perform an experimental evaluate performance of sampling regarding structure preservation. Results confirm our assumption suggest key points for designing new approach named mino-centric (MCGS). approach, triangle-based algorithm cut-point-based proposed efficiently identify structures. A set importance assessment criteria designed guide preservation Three optimization objectives introduced into greedy strategy balance between majority suppress generation series experiments case studies conducted effectiveness MCGS.