P13-1105 treebank with complete trees through direct projection . So we extract projected discrete
P02-1050 results comparing the output of the Direct Projection Algorithm with the gold standard
P12-2010 and recall than the system with direct projection for all of the four relation
D14-1096 data and compare the label from direct projection and from the TnT tagger . The
D09-1086 we evaluate the extent of this direct projection on a small handannotated corpus
N01-1026 significantly outperform the raw direct projections on which they were trained .
P12-2010 two systems constructed by the direct projection ( Kim et al. , 2010 ) and graph-based
P11-1061 well as approaches that rely on direct projections , and bridge the gap between
P06-2039 languages are different , e.g. Direct Projection Algorithm ( Hwa et al. , 2005
P11-1061 LP " model does not outperform direct projection for German and Greek , but performs
D09-1086 - many , they gave a heuristic Direct Projection Algorithm ( DPA ) for resolving
P06-2039 have proposed a scheme called direct projection algorithm ( DPA ) which assumes
P02-1050 described by Melamed ( 1998 ) . The direct projection of English dependencies to Chinese
P06-2039 call this algorithm as pseudo Direct Projection Algorithm ( pDPA ) . The present
P13-1105 as an example , following the Direct Projection Algorithm ( DPA ) ( Hwa et al.
P12-2010 proposed ( Kim et al. , 2010 ) , the direct projection approach is still vulnerable
H01-1035 alignments were manually corrected , direct projection core-tag accuracy increased to
P02-1050 languages other than English . 3.1 The Direct Projection Algorithm The DCA translates
N01-1026 further illustrated that simple direct projection of POS and NP annotations across
H01-1035 tagset granularity of English , direct projection of core POS tags onto French
hide detail