Abstract | ||
---|---|---|
•The novel feature augmentation method, which utilizes the hidden features, the raw features, and zero vectors, is proposed.•The novel hidden feature transformation model is proposed based on the maximum joint probability principle.•With hinge loss function and least square loss function, two semi-supervised classification formulations are proposed. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1016/j.asoc.2017.06.017 | Applied Soft Computing |
Keywords | Field | DocType |
Semi-supervised learning,Cluster assumption,Manifold assumption,Hidden features,Joint probability distribution | Feature vector,Semi-supervised learning,Joint probability distribution,Subspace topology,Pattern recognition,Projection (linear algebra),Robustness (computer science),Orthonormal basis,Artificial intelligence,Machine learning,Mathematics,Manifold | Journal |
Volume | ISSN | Citations |
59 | 1568-4946 | 2 |
PageRank | References | Authors |
0.36 | 29 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Wenlong Hang | 1 | 8 | 0.89 |
Kup-Sze Choi | 2 | 526 | 47.41 |
Shitong Wang | 3 | 1485 | 109.13 |
Pengjiang Qian | 4 | 133 | 11.25 |