id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_4lo326pejbdlzbws57oadv4af4	Nikolaos Polatidis	Privacy-preserving collaborative recommendations based on random perturbations	2017		.pdf	text/html	1398	230	37	Applied Marketing Research Group Bristol Group for Water Research Bristol Inter-disciplinary Group for Education Research Centre for Architecture and Built Environment Research Centre for Fine Print Research Centre for Health and Clinical Research Computer Science Research Centre Digital Cultures Research Centre Environmental Law and Sustainability Research Group Psychological Sciences Research Group Research Group in Mathematics and its Applications Social Science Research Group Polatidis, Nikolaos; Georgiadis, Christos K.; Pimenidis, Elias; Mouratidis, Haralambos In addition, the user ratings utilized in collaborative filtering systems to recommend products or services must be protected. This paper proposes a multi-level privacy-preserving method for collaborative filtering systems by perturbing each rating before it is submitted to the server. Keywords collaborative filtering, random perturbations, multi-level privacy, recommender systems UWE Bristol Research Repository Powered by Worktribe | About UWE Bristol Research Repository Research Centres/Groups Research Centres/Groups Research Centres/Groups HAS Dept of Health & Social Sciences RBI Research & Business Enterprise Service	./cache/work_4lo326pejbdlzbws57oadv4af4.pdf	./txt/work_4lo326pejbdlzbws57oadv4af4.txt