Similarities on Graphs: Kernels versus Proximity Measures
arXiv:1802.06284 · doi:10.1016/j.ejc.2018.02.002
Abstract
We analytically study proximity and distance properties of various kernels and similarity measures on graphs. This helps to understand the mathematical nature of such measures and can potentially be useful for recommending the adoption of specific similarity measures in data analysis.
arXiv comment16 pages
References in corpus (9)
- Communicability in complex networks
- Statistical-mechanical approach to subgraph centrality in complex networks
- Supervised Random Walks: Predicting and Recommending Links in Social Networks
- Developments in the theory of randomized shortest paths with a comparison of graph node distances
- A Class of Graph-Geodetic Distances Generalizing the Shortest-Path and the Resistance Distances
- The Walk Distances in Graphs
- Studying new classes of graph metrics
- The graph bottleneck identity
- Do logarithmic proximity measures outperform plain ones in graph clustering?