Research Article
A Semisupervised Framework for Automatic Image Annotation Based on Graph Embedding and Multiview Nonnegative Matrix Factorization
Table 2
Comparison results with other annotation algorithms.
| Methods | Pre | Rec | F1 | N+ |
| SML | 23 | 29 | 25.7 | 137 | JEC | 27 | 32 | 29.3 | 139 | GS | 30 | 33 | 31.4 | 146 | MRFA | 31 | 36 | 33.3 | 172 | TagProp(ML) | 31 | 37 | 33.7 | 146 | TagProp(ML) | 33 | 42 | 37.0 | 160 | RMLF | 29.7 | 32.6 | 31.1 | - | Merging features | 33 | 40 | 36.5 | - | Scoefficients | 30 | 39 | 34.6 | - | LJNMF(3f’) | 35 | 43 | 39.1 | - | 2PKNN(3f) | 32 | 28 | 30.6 | 177 | SDMKL | 38 | 25 | 30 | 158 | LDMKL | 44 | 29 | 34.9 | 179 | GENMF (3f) | 38 | 39 | 39.2 | 168 |
|
|