Federated cross-view e-commerce recommendation based on feature rescaling
Ruiheng Li, Yuhang Shu, Yue Cao, Yiming Luo, Qiankun Zuo, Xuan Wu, Jiaojiao Yu, Wenxin Zhang · 2024
Read Federated cross-view e-commerce recommendation based on feature rescaling by Ruiheng Li, Yuhang Shu, Yue Cao, Yiming Luo, Qiankun Zuo, Xuan Wu, Jiaojiao Yu, Wenxin Zhang (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11612284.txt.
Description
Eight researchers present a privacy-keeping recommendation method for online shopping, in a 2024 paper. Standard recommenders gather user data in one place, which raises privacy worries. Federated learning trains models on user devices and merges only the settings, yet it struggles with messy, high-dimensional shopping data and noise.
Their framework, Fed-FR-MVD, views data from both the item side and the user side. A rescaling step adjusts features on the fly to sharpen the signal. Tests showed accuracy gains of twelve to eighteen percent over single-view and other multi-view rivals. Performance also held up when noise was added to the data. The authors say the design fills a gap for shopping sites that need good suggestions without pooling raw data. Robustness to noise is a key selling point for real stores.
Catalog record
| CREATOR | Ruiheng Li, Yuhang Shu, Yue Cao, Yiming Luo, Qiankun Zuo, Xuan Wu, Jiaojiao Yu, Wenxin Zhang |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 65 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |