Mixed-order spectral clustering for complex networks

作者：

Highlights：

• Mixed-order spectral clustering to model both second-order and third-order structures simultaneously for complex networks.

• Mixed-order cut criteria for performance enhancement and mixed-order evaluation metrics for structural-level evaluation.

• Extensive experiments on community detection and superpixel segmentation tasks to evaluate the performance of the proposed methods.

摘要

•Mixed-order spectral clustering to model both second-order and third-order structures simultaneously for complex networks.•Mixed-order cut criteria for performance enhancement and mixed-order evaluation metrics for structural-level evaluation.•Extensive experiments on community detection and superpixel segmentation tasks to evaluate the performance of the proposed methods.

论文关键词：Spectral clustering,Higher-order structures,Mixed-order structures

论文评审过程：Received 5 March 2020, Revised 24 October 2020, Accepted 26 March 2021, Available online 6 April 2021, Version of Record 19 April 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.107964