Abstract | ||
---|---|---|
We present an algorithm for multiclass Semi-Supervised learning which is learning from a limited amount of labeled data and plenty of unlabeled data. Existing semi-supervised algorithms use approaches such as one-versus-all to convert the multiclass problem to several binary classification problems which is not optimal. We propose a multiclass semi-supervised boosting algorithm that solves multiclass classification problems directly. The algorithm is based on a novel multiclass loss function consisting of the margin cost on labeled data and two regularization terms on labeled and unlabeled data. Experimental results on a number of UCI datasets show that the proposed algorithm performs better than the state-of-the-art boosting algorithms for multiclass semi-supervised learning. |
Year | DOI | Venue |
---|---|---|
2012 | 10.1109/ICDM.2012.119 | Data Mining |
Keywords | Field | DocType |
learning (artificial intelligence),pattern classification,AdaBoost algorithm,binary classification problem,labeled data,margin cost,multiclass loss function,multiclass semisupervised boosting algorithm,multiclass semisupervised learning,one-versus-all approach,regularization term,unlabeled data,Semi-Supervised Learning,boosting,multiclass classification | Adaboost algorithm,Data mining,Semi-supervised learning,Pattern recognition,Binary classification,Computer science,Regularization (mathematics),Boosting (machine learning),Artificial intelligence,Labeled data,Machine learning,Multiclass classification | Conference |
ISSN | ISBN | Citations |
1550-4786 | 978-1-4673-4649-8 | 6 |
PageRank | References | Authors |
0.42 | 10 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jafar Tanha | 1 | 50 | 4.17 |
Maarten van Someren | 2 | 402 | 48.51 |
Hamideh Afsarmanesh | 3 | 1890 | 291.69 |