E14-1020 First , we introduce a chart based surface realisation algorithm which supports the
E09-1097 precedence order , bypassing the surface realisation stage . 7 Conclusions In this
E12-1078 sentence-external context in a surface realisation task . The sentence-external
E12-1078 correspond to the constituents that our surface realisation model has to put in the right
E14-1020 Second , we show that grammar based surface realisation better supports the generation
D15-1224 selection is more important than surface realisation for REG . Secondly , semantic
D12-1085 addressed in the 2011 Shared Task on Surface Realisation ( Belz et al. , 2011 ) . Here
E14-1020 Random Fields to find the best surface realisation from a semantic tree . They show
E09-1014 present a human-based evaluation of surface realisation alterna - tives . We examine
D14-1007 mechanisms are isolated , for which the surface realisations sometimes may sound stiff . This
E12-1078 statistical ranking model for surface realisation of German ( Section 4 ) . As
E09-1014 presented between 4 and 8 alternative surface realisations for an input f-structure , as
C04-1087 frequently denoted by different surface realisations of preferred terms , which we
C94-2192 have not , h - ing to do with the surface realisations of the relations . On the other
E09-1014 which 96 % can be associated with surface realisations by the grammar . F-structures
C02-2011 much connected to morphosyntactic surface realisation as to syntactico-semantic categories
D12-1085 systems . In the 2011 Shared Task on Surface Realisation ( Belz et al. , 2011 ) , the
E12-1078 entities . Other work on German surface realisation has highlighted the role of the
E09-1000 Lapata Human Evaluation of a German Surface Realisation Ranker Bruno Cartoni Mathias
E06-2020 2000 ) . The microplanning and surface realisation modules from the Sumtime project
hide detail