PDMFRec: A Decentralised Matrix Factorisation with Tunable User-centric Privacy
|Title:||PDMFRec: A Decentralised Matrix Factorisation with Tunable User-centric Privacy||Authors:||Duriakova, Erika; Tragos, Elias; Smyth, Barry; Hurley, Neil J.; Peña, Francisco; Symeonidis, Panagiotis; Geraci, James; Lawlor, Aonghus||Permanent link:||http://hdl.handle.net/10197/11360||Date:||19-Sep-2019||Online since:||2020-04-30T15:04:50Z||Abstract:||Conventional approaches to matrix factorisation (MF) typically rely on a centralised collection of user data for building a MF model. This approach introduces an increased risk when it comes to user privacy. In this short paper we propose an alternative, user-centric, privacy enhanced, decentralised approach to MF. Our method pushes the computation of the recommendation model to the user’s device, and eliminates the need to exchange sensitive personal information; instead only the loss gradients of local (device-based) MF models need to be shared. Moreover, users can select the amount and type of information to be shared, for enhanced privacy. We demonstrate the effectiveness of this approach by considering different levels of user privacy in comparison with state-of-the-art alternatives.||Funding Details:||Science Foundation Ireland||metadata.dc.description.othersponsorship:||Insight Research Centre
|Type of material:||Conference Publication||Publisher:||ACM||Copyright (published version):||2019 ACM||Keywords:||Matrix factorisation; Decentralised matrix factorisation; Privacy aware; Rating prediction||DOI:||10.1145/3298689.3347035||Other versions:||https://recsys.acm.org/recsys19/||Language:||en||Status of Item:||Peer reviewed||Conference Details:||The 13th ACM Conference on Recommender Systems (RecSys'19), Copenhagen, Denmark, 16-20 September 2019||ISBN:||978-1-4503-6243-6|
|Appears in Collections:||Insight Research Collection|
Show full item record
This item is available under the Attribution-NonCommercial-NoDerivs 3.0 Ireland. No item may be reproduced for commercial purposes. For other possible restrictions on use please refer to the publisher's URL where this is made available, or to notes contained in the item itself. Other terms may apply.