Abstract
Most semi-supervised learning methods are based on extending existing supervised or unsupervised techniques by incorporating additional information from unlabeled or labeled data. Unlabeled instances help in learning statistical models that fully describe the global property of our data, whereas labeled instances make learned knowledge more human-interpretable. In this paper we present a novel way of extending conventional non-negativematrix factorization (NMF) and probabilistic latent semantic analysis (pLSA) to semi-supervised versions by incorporating label information for learning semantics. The proposed algorithm consists of two steps, first acquiring prior bases representing some classes from labeled data and second utilizing them to guide the learning of final bases that are semantically interpretable.
| Original language | English |
|---|---|
| Pages (from-to) | 366-374 |
| Number of pages | 9 |
| Journal | Journal of Advanced Computational Intelligence and Intelligent Informatics |
| Volume | 18 |
| Issue number | 3 |
| Publication status | Published - 2014 |
Fingerprint
Keywords
- Non-negative matrix factorization
- Probabilistic latent semantic analysis
- Semi-supervised learning
ASJC Scopus subject areas
- Artificial Intelligence
- Computer Vision and Pattern Recognition
- Human-Computer Interaction
Cite this
Hierarchical semi-supervised factorization for learning the semantics. / Shen, Bin; Makhambetov, Olzhas.
In: Journal of Advanced Computational Intelligence and Intelligent Informatics, Vol. 18, No. 3, 2014, p. 366-374.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Hierarchical semi-supervised factorization for learning the semantics
AU - Shen, Bin
AU - Makhambetov, Olzhas
PY - 2014
Y1 - 2014
N2 - Most semi-supervised learning methods are based on extending existing supervised or unsupervised techniques by incorporating additional information from unlabeled or labeled data. Unlabeled instances help in learning statistical models that fully describe the global property of our data, whereas labeled instances make learned knowledge more human-interpretable. In this paper we present a novel way of extending conventional non-negativematrix factorization (NMF) and probabilistic latent semantic analysis (pLSA) to semi-supervised versions by incorporating label information for learning semantics. The proposed algorithm consists of two steps, first acquiring prior bases representing some classes from labeled data and second utilizing them to guide the learning of final bases that are semantically interpretable.
AB - Most semi-supervised learning methods are based on extending existing supervised or unsupervised techniques by incorporating additional information from unlabeled or labeled data. Unlabeled instances help in learning statistical models that fully describe the global property of our data, whereas labeled instances make learned knowledge more human-interpretable. In this paper we present a novel way of extending conventional non-negativematrix factorization (NMF) and probabilistic latent semantic analysis (pLSA) to semi-supervised versions by incorporating label information for learning semantics. The proposed algorithm consists of two steps, first acquiring prior bases representing some classes from labeled data and second utilizing them to guide the learning of final bases that are semantically interpretable.
KW - Non-negative matrix factorization
KW - Probabilistic latent semantic analysis
KW - Semi-supervised learning
UR - http://www.scopus.com/inward/record.url?scp=84901017341&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84901017341&partnerID=8YFLogxK
M3 - Article
AN - SCOPUS:84901017341
VL - 18
SP - 366
EP - 374
JO - Journal of Advanced Computational Intelligence and Intelligent Informatics
JF - Journal of Advanced Computational Intelligence and Intelligent Informatics
SN - 1343-0130
IS - 3
ER -