MVDF-RSC: Multi-view data fusion via robust spectral clustering for geo-tagged image tagging
作者：
Highlights：
• We propose a new robust multi-view image tagging method via the MCC-based framework.
• We use a diversity regularization term to promote complementary information.
• The proposed clustering method finds clusters with no additional clustering step.
• A compelling fusion technique with the combination of early and late fusion is used.
• Geographical information is used to enhance the performance of the model.
摘要
•We propose a new robust multi-view image tagging method via the MCC-based framework.•We use a diversity regularization term to promote complementary information.•The proposed clustering method finds clusters with no additional clustering step.•A compelling fusion technique with the combination of early and late fusion is used.•Geographical information is used to enhance the performance of the model.
论文关键词：Multi-view spectral clustering,Image annotation,Geo-tagged photos,Image tagging,Recommender systems,Geographical information
论文评审过程：Received 23 November 2020, Revised 14 January 2021, Accepted 23 January 2021, Available online 4 February 2021, Version of Record 20 February 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.114657
