Seeded random walk for multi-view semi-supervised classification

作者：

Highlights：

• Propose an efficient random walk based multi-view semi-supervised learning scheme.

• Develop an effective classification method for a limited number of labeled samples.

• Transform classification problem as an estimation of overall arrival probability.

• Prove that the proposed method is an extension to well-known manifold embedding.

摘要

•Propose an efficient random walk based multi-view semi-supervised learning scheme.•Develop an effective classification method for a limited number of labeled samples.•Transform classification problem as an estimation of overall arrival probability.•Prove that the proposed method is an extension to well-known manifold embedding.

论文关键词：Machine learning,Multi-view fusion,Semi-supervised classification,Random walk,Arrival probability,Reward matrix

论文评审过程：Received 10 December 2020, Revised 24 February 2021, Accepted 31 March 2021, Available online 7 April 2021, Version of Record 15 April 2021.

论文官网地址：https://doi.org/10.1016/j.knosys.2021.107016