Research Article
A Novel Approach of Feature Space Reconstruction with Three-Way Decisions for Long-Tailed Text Classification
Table 2
The numbers of the long-tailed text dataset used in our paper.
| No. | Class | Class label | Samples |
| 1 | Head | C34-Economy | 3201 | 2 | Head | C39-Sports | 2507 | 3 | Head | C38-Politics | 1854 | 4 | Head | C3-Art | 1282 | 5 | Tail | C11-Space | 582 | 6 | Tail | C32-Agriculture | 348 | 7 | Tail | C37-Military | 150 | 8 | Tail | C19-Computer | 120 | 9 | Tail | C36-Medical | 104 | 10 | Tail | C35-Law | 103 | 11 | Tail | C31-Environment | 89 | 12 | Tail | C29-Transport | 67 | 13 | Tail | C23-Mine | 67 | 14 | Tail | C15-Energy | 65 | 15 | Tail | C17-Communication | 55 | 16 | Tail | C16-Electronics | 32 |
|
|