Graph kernels based on linear patterns: Theoretical and experimental comparisons

作者：

Highlights：

• Graph kernels are powerful to bridge the gap between machine learning and graphs.

• Graph kernels based on linear patterns can achieve good performances.

• Graph kernels should be chosen carefully according to the types of graph datasets.

摘要

•Graph kernels are powerful to bridge the gap between machine learning and graphs.•Graph kernels based on linear patterns can achieve good performances.•Graph kernels should be chosen carefully according to the types of graph datasets.

论文关键词：Graph kernels,Walks,Paths,Kernel methods,Graph representation

论文评审过程：Received 4 March 2020, Revised 13 October 2021, Accepted 13 October 2021, Available online 30 October 2021, Version of Record 8 November 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.116095