Research Area:  Data Mining

Abstract:

Recommender systems based on methods such as collaborative and content-based filtering rely on extensive user profiles and item descriptors as well as on an extensive history of user preferences. Such methods face a number of challenges; including the cold-start problem in systems characterized by irregular usage, privacy concerns, and contexts where the range of indicators representing user interests is limited. We describe a recommender algorithm that builds a model of collective preferences independently of personal user interests and does not require a complex system of ratings. The performance of the algorithm is analyzed on a large transactional data set generated by a real-world dietary intake recall system.

Keywords:  

Author(s) Name:  Timur Osadchiy,Ivan Poliakov,Patrick Olivier,Maisie Rowland and Emma Foster

Publisher name:  ELSEVIER

DOI:  10.1016/j.eswa.2018.07.077

Volume Information:  Volume 115, January 2019, Pages 535-542

Paper Link:   https://www.sciencedirect.com/science/article/pii/S095741741830441X