| P06-2010 | convolution tree kernel outperforms the | PAF kernel | method . Although the hybrid |
| P06-2010 | convolution tree kernel outperforms the | PAF kernel | . It empirically demonstrates |
| P06-2010 | kernel is more effective than | PAF kernel | for SRL . However , our hybrid |
| P06-2010 | kernel machine outperforms the | PAF kernel | in the development sets of CoNLL-2005 |
| P06-2010 | well . Affected by this , the | PAF kernel | can not perform well , either |
| P06-2010 | of the distinction between the | PAF kernel | and our kernel . In the PAF kernel |
| P06-2010 | kernel plays the main role in | PAF kernel | computation , as shown in Figure |
| P06-2010 | Convolution Tree Kernels In the | PAF kernel | , the feature spaces are considered |
| P06-2010 | kernel and our kernel . In the | PAF kernel | , the tree structures are equal |
| P06-2010 | nel . Figure 2 illustrates the | PAF kernel | feature space of the predicate |
| P06-2010 | kernel , Moschitti ( 2004 ) 's | PAF kernel | , standard flat features with |
