Mixture distribution modeling for scalable graph-based semi-supervised learning

作者：

Highlights：

• We propose a novel scalable SSL, which overcomes the limitations of existing methods.

• We extend our method to a more general framework for anchor-based scalable SSL.

• The experimental results demonstrate the superiority of the proposed model.

摘要

•We propose a novel scalable SSL, which overcomes the limitations of existing methods.•We extend our method to a more general framework for anchor-based scalable SSL.•The experimental results demonstrate the superiority of the proposed model.

论文关键词：Semi-supervised Learning,Graph-based Learning,Mixture Distribution Modeling

论文评审过程：Received 26 November 2019, Revised 27 March 2020, Accepted 24 April 2020, Available online 5 May 2020, Version of Record 8 May 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2020.105974