Reference : Collaborative filtering: Scalable approaches using restricted Boltzmann machines

	Document type :	Dissertations and theses : Master's dissertation
	Discipline(s) :	Engineering, computing & technology : Computer science
	To cite this reference:	http://hdl.handle.net/2268/74400

Title :	Collaborative filtering: Scalable approaches using restricted Boltzmann machines
Language :	English
Author, co-author :	Louppe, Gilles [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation >]
Publication date :	29-Jun-2010
Institution :	Université de Liège, ​Liège, ​​Belgique
Name of the degree :	Master en sciences informatiques, à finalité approfondie
Page number :	95
Superviser :	Geurts, Pierre
President of the jury :	Boigelot, Bernard
Member of the jury :	Sepulchre, Rodolphe
	Wehenkel, Louis
	Leduc, Guy
Keywords :	[en] collaborative filtering ; restricted boltzmann machines ; machine learning
Abstract :	[en] Parallel to the growth of electronic commerce, recommender systems have become a very active area of research, both in the industry and in the academic world. The goal of these systems is to make automatic but personal recommendations when customers are overwhelmed with thousands of possibilities and do not know what to look for. In that context, the object of this work is threefold. The first part consists in a survey of recommendation algorithms and emphasizes on a class of algorithms known as collaborative filtering algorithms. The second part consists in studying in more depth a specific model of neural networks known as restricted Boltzmann machines. That model is then experimentaly and extensively examined on a recommendation problem. The third part of this work focuses on how restricted Boltzmann machines can be made more scalable. Three different and original approaches are proposed and studied. In the first approach, we revisit the learning and test algorithms of restricted Boltzmann machines in the context of shared-memory architectures. In the second approach, we propose to reformulate these algorithms as MapReduce tasks. Finally, in the third method, ensemble of RBMs are investigated. The best and the more promising results are obtained with the MapReduce approach.
Target :	Researchers
Permalink :	http://hdl.handle.net/2268/74400

	File(s) associated to this reference
	Fulltext file(s): File Commentary Version Size Access Open access glouppe10-msc-thesis.pdf Publisher postprint 5.3 MB View/Open

All documents in ORBi are protected by a user license.

 

© ULg Library Network |  Feedback |  ORBi legal notices

Dspace v1.5