Adaptive hypergraph learning with multi-stage optimizations for image and tag recommendation

作者：

Highlights：

• A learning scheme for hypergraph ranking based on multiple optimizations is proposed.

• The proposed scheme optimizes hypergraph structure, hyperedge weights, and hypergraph ranking vectors.

• The proposed optimizations are solved analytically and full derivations are provided.

• Experiments are conducted on two public datasets for image and tag recommendation.

摘要

•A learning scheme for hypergraph ranking based on multiple optimizations is proposed.•The proposed scheme optimizes hypergraph structure, hyperedge weights, and hypergraph ranking vectors.•The proposed optimizations are solved analytically and full derivations are provided.•Experiments are conducted on two public datasets for image and tag recommendation.

论文关键词：Recommendation system,Hypergraphs,Multi-stage optimizations,Structure learning,Weight adaptation

论文评审过程：Received 5 February 2021, Revised 28 May 2021, Accepted 22 June 2021, Available online 28 June 2021, Version of Record 5 July 2021.

论文官网地址：https://doi.org/10.1016/j.image.2021.116367