Abstract | ||
---|---|---|
Graph based semi-supervised learning methods (SSL) implicitly assume that the intrinsic geometry of the data points can be fully specified by an Euclidean distance based local neighborhood graph, however, this assumption may not always be necessarily true. To overcome this problem, we propose to apply locally linear embedding (LLE) method to characterize the geometric structure of the data points; besides this, the embedding process is performed in the kernel induced feature space rather than the original input space. After embedding, the proposed transductive learning method predicts the labels of the unlabeled data within the regularization framework. Experimental results on image retrieval and pattern recognition verify the performance of the proposed approach. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1109/ICPR.2008.4761127 | ICPR |
Keywords | Field | DocType |
local neighborhood graph,locally linear embedding,semi supervised learning,learning (artificial intelligence),graph regularization framework,euclidean distance,transductive learning method,computational geometry,intrinsic data point geometry,kernel space,learning artificial intelligence,feature space,kernel,transductive learning,pattern recognition,classification algorithms,geometry,manifolds,image retrieval | Transduction (machine learning),Kernel (linear algebra),Feature vector,Semi-supervised learning,Embedding,Pattern recognition,Kernel embedding of distributions,Computer science,Euclidean distance,Artificial intelligence,Statistical classification | Conference |
ISSN | ISBN | Citations |
1051-4651 E-ISBN : 978-1-4244-2175-6 | 978-1-4244-2175-6 | 0 |
PageRank | References | Authors |
0.34 | 10 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Rujie Liu | 1 | 147 | 15.49 |
Yuehong Wang | 2 | 72 | 4.66 |
Takayuki Baba | 3 | 77 | 8.19 |
Daiki Masumoto | 4 | 76 | 6.33 |