Leveraging unlabeled data with a probabilistic graphical model

Details Leveraging unlabeled data with a probabilistic graphical model Leveraging unlabeled data with a probabilistic graphical model

2011-05-03

2011-05-03

Jackson, Jakieda R (Department: 2626)

Data processing: speech signal processing, linguistics, language

Linguistics

Natural language

Reexamination Certificate

active

07937264

ABSTRACT:
A general probabilistic formulation referred to as ‘Conditional Harmonic Mixing’ is provided, in which links between classification nodes are directed, a conditional probability matrix is associated with each link, and where the numbers of classes can vary from node to node. A posterior class probability at each node is updated by minimizing a divergence between its distribution and that predicted by its neighbors. For arbitrary graphs, as long as each unlabeled point is reachable from at least one training point, a solution generally always exists, is unique, and can be found by solving a sparse linear system iteratively. In one aspect, an automated data classification system is provided. The system includes a data set having at least one labeled category node in the data set. A semi-supervised learning component employs directed arcs to determine the label of at least one other unlabeled category node in the data set.

REFERENCES:
patent: 6154736 (2000-11-01), Chickering et al.
patent: 7310624 (2007-12-01), Aggarwal et al.
patent: 2002/0010691 (2002-01-01), Chen
patent: 2002/0107853 (2002-08-01), Hofmann et al.
patent: 2003/0140020 (2003-07-01), Chen et al.
patent: 2004/0181441 (2004-09-01), Fung et al.
patent: 2004/0202987 (2004-10-01), Scheuring et al.
patent: 2004/0220892 (2004-11-01), Cohen et al.
patent: 2005/0055209 (2005-03-01), Epstein et al.
patent: 2006/0015341 (2006-01-01), Baker
patent: 2006/0026112 (2006-02-01), Naphade et al.
patent: 2006/0093208 (2006-05-01), Li et al.
patent: 2006/0167655 (2006-07-01), Barrow et al.
patent: 2006/0235812 (2006-10-01), Rifkin et al.
patent: 2006/0288275 (2006-12-01), Chidlovskii et al.
Golub et al., Matrix Computations, Johns Hopkins, third edition, 1996, 694 pages, Baltimore, Maryland, USA.
Doyle et al., Random Walks and Electric Networks., The Mathematical Associatino of America, 1984.
Zhu et al., Semi-supervised learning using Gaussian fields and harmonic functions, In Proceedings of the 20th ICML, 2003,Washington D.C.
Dengyong Zhou, Learning with local and global consistency, Advances in Neural Information Processing Systems, MIT Press, 2004, pp. 321-328, vol. 16.
Thorsten Joachims, Transductive Inference for Text Classification Using Support Vector Machines., Proceedings of ICML-99, 1999, pp. 200-209, San Francisco, CA, USA.
Zhu et al., Semi-supervised learning using gaussian fields and harmonic functions, Proceedings of the 20th ICML, 2003, 8 pages, Washington D.C., USA.

Affiliated with

Also associated with

Rating

Leveraging unlabeled data with a probabilistic graphical model does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Leveraging unlabeled data with a probabilistic graphical model, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Leveraging unlabeled data with a probabilistic graphical model will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2667688