Document Detail


CoTrade: Confident Co-Training With Data Editing.
MedLine Citation:
PMID:  21708503     Owner:  NLM     Status:  Publisher    
Abstract/OtherAbstract:
Co-training is one of the major semi-supervised learning paradigms that iteratively trains two classifiers on two different views, and uses the predictions of either classifier on the unlabeled examples to augment the training set of the other. During the co-training process, especially in initial rounds when the classifiers have only mediocre accuracy, it is quite possible that one classifier will receive labels on unlabeled examples erroneously predicted by the other classifier. Therefore, the performance of co-training style algorithms is usually unstable. In this paper, the problem of how to reliably communicate labeling information between different views is addressed by a novel co-training algorithm named CoTrade. In each labeling round, CoTrade carries out the label communication process in two steps. First, confidence of either classifier's predictions on unlabeled examples is explicitly estimated based on specific data editing techniques. Secondly, a number of predicted labels with higher confidence of either classifier are passed to the other one, where certain constraints are imposed to avoid introducing undesirable classification noise. Experiments on several real-world datasets across three domains show that CoTrade can effectively exploit unlabeled data to achieve better generalization performance.
Authors:
Min-Ling Zhang; Zhi-Hua Zhou
Related Documents :
21881513 - The use of a cystic fibrosis patient registry to assess outcomes and improve cystic fib...
8211483 - Risk-taking behaviour of cape peninsula high-school students. part ii. suicidal behaviour.
19171823 - Education research: neurology residency training in the new millennium.
Publication Detail:
Type:  JOURNAL ARTICLE     Date:  2011-6-23
Journal Detail:
Title:  IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society     Volume:  -     ISSN:  1941-0492     ISO Abbreviation:  -     Publication Date:  2011 Jun 
Date Detail:
Created Date:  2011-6-28     Completed Date:  -     Revised Date:  -    
Medline Journal Info:
Nlm Unique ID:  9890044     Medline TA:  IEEE Trans Syst Man Cybern B Cybern     Country:  -    
Other Details:
Languages:  ENG     Pagination:  -     Citation Subset:  -    
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  A Bayesian theoretic approach to multi-scale complex phase order representations.
Next Document:  Fast and Accurate Fingerprint Indexing Based on Ridge Orientation and Frequency.