Conference Proceeding

Incremental Non-negative Matrix Factorization for Dynamic Background Modelling.

01/2007; In proceeding of: Pattern Recognition in Information Systems, Proceedings of the 7th International Workshop on Pattern Recognition in Information Systems, PRIS 2007, In conjunction with ICEIS 2007, Funchal, Madeira, Portugal, June 2007
Source: DBLP
0 0
  • [show abstract] [hide abstract]
    ABSTRACT: PLSA which was originally introduced in text analysis area, has been extended to predict user ratings in the collaborative filtering context, known as Triadic PLSA (TPLSA). It is a promising recommender technique but the computational cost is a bottleneck for huge data set. We design a incremental learning scheme for TPLSA for collaborative filtering task that could make forced prediction and free prediction as well. Our incremental implementation is the first of its kind in the probabilistic model based collaborative filtering area, to our best knowledge. Its effectiveness is validated by experiments designed for both rating-based and ranking-based collaborative filtering.
    Active Media Technology, 5th International Conference, AMT 2009, Beijing, China, October 22-24, 2009. Proceedings; 01/2009
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: Despite its relative novelty, non-negative matrix factorization (NMF) method knewahugeinterestfromthescientificcommunity,duetoitssimplicityandintuitivedecom- position. Plenty of applications benefited from it, including image processing (face, medical, etc.), audio data processing or text mining and decomposition. This paper briefly describes the underlaying mathematical NMF theory along with some extensions. Several relevant applications from different scientific areas are also presented. NMF shortcomings and con- clusions are considered.

Full-text (2 Sources)

Available from
Sep 23, 2013