Sparse fuzzy classification for profiling online users and relevant user-generated content

作者：

Highlights：

• This study investigates a distinctive type of online crowd-sourced activities.

• Elementary and aggregated fuzzy sets are introduced to fuzzify samples.

• Sparsity-enhanced regularization is introduced for classification.

• Experiments show superior performance compared to state-of-the-art approaches.

摘要

•This study investigates a distinctive type of online crowd-sourced activities.•Elementary and aggregated fuzzy sets are introduced to fuzzify samples.•Sparsity-enhanced regularization is introduced for classification.•Experiments show superior performance compared to state-of-the-art approaches.

论文关键词：User online behavior,User profile,Fuzzy set,Sparse classification,Crowd-sourced media

论文评审过程：Received 23 September 2020, Revised 13 February 2021, Accepted 31 December 2021, Available online 19 January 2022, Version of Record 21 January 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.116497