Research Article
A Compressive Sensing Model for Speeding Up Text Classification
Table 5
Average accuracies of all classifiers for binary and multiclass classification datasets when using different DR methods.
| DR method | Subrate R | 0.1 | 0.2 | 0.3 | 0.4 | 0.5 | 0.6 |
| Binary classification | PCA | 0.6221 | 0.6236 | 0.6206 | 0.6154 | 0.6222 | 0.6091 | ICA | 0.5754 | 0.5830 | 0.5862 | 0.5974 | 0.5903 | 0.6009 | NMF | 0.5926 | 0.6127 | 0.6193 | 0.6067 | 0.6157 | 0.6000 | CS | 0.6670 | 0.6751 | 0.6829 | 0.6793 | 0.6824 | 0.6827 |
| Multiclass classification | PCA | 0.7253 | 0.7213 | 0.7019 | 0.6845 | 0.6822 | 0.6726 | ICA | 0.4938 | 0.5170 | 0.5305 | 0.5448 | 0.5455 | 0.5479 | NMF | 0.7112 | 0.7080 | 0.7123 | 0.7123 | 0.7096 | 0.7063 | CS | 0.7960 | 0.8074 | 0.8116 | 0.8143 | 0.8156 | 0.8171 |
|
|
Note that SRM in CS is Block DCT.
|