tech,7-1-P01-1056,ak automatically training modules of a <term> natural language generator </term> have recently been proposed , but
measure(ment),22-1-P01-1056,ak fundamental concern is whether the <term> quality </term> of <term> utterances </term> produced
other,24-1-P01-1056,ak whether the <term> quality </term> of <term> utterances </term> produced with trainable components
tech,32-1-P01-1056,ak trainable components can compete with <term> hand-crafted template-based or rule-based approaches </term> . In this paper We experimentally
tech,7-2-P01-1056,ak paper We experimentally evaluate a <term> trainable sentence planner </term> for a <term> spoken dialogue system
tech,12-2-P01-1056,ak trainable sentence planner </term> for a <term> spoken dialogue system </term> by eliciting <term> subjective human
other,17-2-P01-1056,ak dialogue system </term> by eliciting <term> subjective human judgments </term> . In order to perform an exhaustive
tech,12-3-P01-1056,ak exhaustive comparison , we also evaluate a <term> hand-crafted template-based generation component </term> , two <term> rule-based sentence planners
tech,18-3-P01-1056,ak template-based generation component </term> , two <term> rule-based sentence planners </term> , and two <term> baseline sentence
tech,24-3-P01-1056,ak sentence planners </term> , and two <term> baseline sentence planners </term> . We show that the <term> trainable
tech,4-4-P01-1056,ak planners </term> . We show that the <term> trainable sentence planner </term> performs better than the <term> rule-based
tech,11-4-P01-1056,ak planner </term> performs better than the <term> rule-based systems </term> and the <term> baselines </term> , and
tech,15-4-P01-1056,ak <term> rule-based systems </term> and the <term> baselines </term> , and as well as the <term> hand-crafted
tech,22-4-P01-1056,ak baselines </term> , and as well as the <term> hand-crafted system </term> . We describe a set of <term> supervised
skrij podrobnosti