A survey of collaborative filtering techniques
Reads0
Chats0
TLDR
From basic techniques to the state-of-the-art, this paper attempts to present a comprehensive survey for CF techniques, which can be served as a roadmap for research and practice in this area.Abstract:
As one of the most successful approaches to building recommender systems, collaborative filtering (CF) uses the known preferences of a group of users to make recommendations or predictions of the unknown preferences for other users. In this paper, we first introduce CF tasks and their main challenges, such as data sparsity, scalability, synonymy, gray sheep, shilling attacks, privacy protection, etc., and their possible solutions. We then present three main categories of CF techniques: memory-based, modelbased, and hybrid CF algorithms (that combine CF with other recommendation techniques), with examples for representative algorithms of each category, and analysis of their predictive performance and their ability to address the challenges. From basic techniques to the state-of-the-art, we attempt to present a comprehensive survey for CF techniques, which can be served as a roadmap for research and practice in this area.read more
Citations
More filters
Journal ArticleDOI
Recommender systems survey
TL;DR: An overview of recommender systems as well as collaborative filtering methods and algorithms is provided, which explains their evolution, provides an original classification for these systems, identifies areas of future implementation and develops certain areas selected for past, present or future importance.
Journal ArticleDOI
Link prediction in complex networks: A survey
TL;DR: Recent progress about link prediction algorithms is summarized, emphasizing on the contributions from physical perspectives and approaches, such as the random-walk-based methods and the maximum likelihood methods.
Proceedings ArticleDOI
Deep Neural Networks for YouTube Recommendations
TL;DR: This paper details a deep candidate generation model and then describes a separate deep ranking model and provides practical lessons and insights derived from designing, iterating and maintaining a massive recommendation system with enormous user-facing impact.
Journal ArticleDOI
TRY - a global database of plant traits
Jens Kattge,Sandra Díaz,Sandra Lavorel,Iain Colin Prentice,Paul Leadley,Gerhard Bönisch,Eric Garnier,Mark Westoby,Peter B. Reich,Peter B. Reich,Ian J. Wright,Johannes H. C. Cornelissen,Cyrille Violle,Sandy P. Harrison,P.M. van Bodegom,Markus Reichstein,Brian J. Enquist,Nadejda A. Soudzilovskaia,David D. Ackerly,Madhur Anand,Owen K. Atkin,Michael Bahn,Timothy R. Baker,Dennis D. Baldocchi,Renée M. Bekker,Carolina C. Blanco,Benjamin Blonder,William J. Bond,Ross A. Bradstock,Daniel E. Bunker,Fernando Casanoves,Jeannine Cavender-Bares,Jeffrey Q. Chambers,F. S. Chapin,Jérôme Chave,David A. Coomes,William K. Cornwell,Joseph M. Craine,B. H. Dobrin,Leandro da Silva Duarte,Walter Durka,James J. Elser,Gerd Esser,Marc Estiarte,William F. Fagan,Jingyun Fang,Fernando Fernández-Méndez,Alessandra Fidelis,Bryan Finegan,Olivier Flores,H. Ford,Dorothea Frank,Grégoire T. Freschet,Nikolaos M. Fyllas,Rachael V. Gallagher,Walton A. Green,Alvaro G. Gutiérrez,Thomas Hickler,Steven I. Higgins,John G. Hodgson,Adel Jalili,Steven Jansen,Carlos Alfredo Joly,Andrew J. Kerkhoff,Don Kirkup,Kaoru Kitajima,Michael Kleyer,Stefan Klotz,Johannes M. H. Knops,Koen Kramer,Ingolf Kühn,Hiroko Kurokawa,Daniel C. Laughlin,Tali D. Lee,Michelle R. Leishman,Frederic Lens,Tanja Lenz,Simon L. Lewis,Jon Lloyd,Jon Lloyd,Joan Llusià,Frédérique Louault,Siyan Ma,Miguel D. Mahecha,Peter Manning,Tara Joy Massad,Belinda E. Medlyn,Julie Messier,Angela T. Moles,Sandra Cristina Müller,Karin Nadrowski,Shahid Naeem,Ülo Niinemets,S. Nöllert,A. Nüske,Romà Ogaya,Jacek Oleksyn,Vladimir G. Onipchenko,Yusuke Onoda,Jenny C. Ordoñez,Gerhard E. Overbeck,Wim A. Ozinga,Sandra Patiño,Susana Paula,Juli G. Pausas,Josep Peñuelas,Oliver L. Phillips,Valério D. Pillar,Hendrik Poorter,Lourens Poorter,Peter Poschlod,Andreas Prinzing,Raphaël Proulx,Anja Rammig,Sabine Reinsch,Björn Reu,Lawren Sack,Beatriz Salgado-Negret,Jordi Sardans,Satomi Shiodera,Bill Shipley,Andrew Siefert,Enio E. Sosinski,Jean-François Soussana,Emily Swaine,Nathan G. Swenson,Ken Thompson,Peter E. Thornton,Matthew S. Waldram,Evan Weiher,Michael T. White,S. White,S. J. Wright,Benjamin Yguel,Sönke Zaehle,Amy E. Zanne,Christian Wirth +136 more
TL;DR: TRY as discussed by the authors is a global database of plant traits, including morphological, anatomical, physiological, biochemical and phenological characteristics of plants and their organs, which can be used for a wide range of research from evolutionary biology, community and functional ecology to biogeography.
References
More filters
Journal ArticleDOI
A personalised TV listings service for the digital TV age
Barry Smyth,Paul Cotter +1 more
TL;DR: The development of the PTV (Personalised Television Listings—http://www.ptv.ie) system is described which tackles the information overload associated with modern TV listings data, by providing an Internet-based personalised listings service.
Proceedings ArticleDOI
Investigation of various matrix factorization methods for large recommender systems
TL;DR: An incremental variant of MF is described that efficiently handles new users/ratings, which is crucial in a real-life recommender system and a momentum-based MF approach is introduced that approximates the features by using positive values for either users or items.
Proceedings Article
A Maximum Entropy Approach to Collaborative Filtering in Dynamic, Sparse, High-Dimensional Domains
TL;DR: This work develops a maximum entropy (maxent) approach to generating recommendations in the context of a user's current navigation stream, suitable for environments where data is sparse, high-dimensional, and dynamic— conditions typical of many recommendation applications.
Journal ArticleDOI
Structural Extension to Logistic Regression: Discriminative Parameter Learning of Belief Net Classifiers
TL;DR: In this article, the authors propose a discriminative parameter learning approach to find the BN that maximizes a different objective function, i.e., likelihood, rather than classification accuracy.
Journal ArticleDOI
An electronic infrastructure for a virtual university
TL;DR: The Electronic Education Environment, or E3, an infrastructure developed at The University of Texas at Austin to support processes in a virtual university (VU), focuses on developing skills and expertise by mass customizing content on demand rather than providing terminal degree programs with homogeneous and predetermined curricula.