Paper: Shallow Semantic Parsing for Spoken Language Understanding

ACL ID N09-2022
Title Shallow Semantic Parsing for Spoken Language Understanding
Venue Human Language Technologies
Session Short Paper
Year 2009

Most Spoken Dialog Systems are based on speech grammars and frame/slot semantics. The semantic descriptions of input utterances are usually defined ad-hoc with no ability to generalize beyond the target application do- main or to learn from annotated corpora. The approach we propose in this paper exploits machine learning of frame semantics, bor- rowing its theoretical model from computa- tional linguistics. While traditional automatic Semantic Role Labeling approaches on writ- ten texts may not perform as well on spo- ken dialogs, we show successful experiments on such porting. Hence, we design and eval- uate automatic FrameNet-based parsers both for English written texts and for Italian dia- log utterances. The results show that disflu- encies of dialog data do not severely hurt per- for...