Research Article
Enhancing Business Intelligence by Means of Suggestive Reviews
Table 1
Performance numbers on LRChi2 over all feature construction approaches for detecting suggestive reviews.
| Feature construction | Precision | Recall | |
| Bow | 0.8806 | 0.7603 | 0.8129 | Bow + Cmp | 0.8797 | 0.7539 | 0.8093 | Bow + Sug | 0.8715 | 0.7668 | 0.8124 | Bow + Cmp + Sug | 0.8796 | 0.7733 | 0.8193 | Bow + Cmp+ Sug + Surf | 0.859 | 0.7572 | 0.8008 | Sug | 0.899 | 0.6535 | 0.7534 | Mix | 0.8388 | 0.7671 | 0.7975 | Sug + Cmp | 0.9126 | 0.6537 | 0.7568 | Sug + Cmp + Surf | 0.8937 | 0.6699 | 0.7586 | Mix + Sug + Cmp | 0.8271 | 0.78 | 0.8005 | Mix + Surf + Sug + Cmp | 0.8309 | 0.7703 | 0.7972 |
|
|