tech,3-1-P03-1033,bq effective . We address appropriate <term> user modeling </term> in order to generate <term> cooperative
other,9-1-P03-1033,bq modeling </term> in order to generate <term> cooperative responses </term> to each <term> user </term> in <term> spoken
other,13-1-P03-1033,bq cooperative responses </term> to each <term> user </term> in <term> spoken dialogue systems </term>
tech,15-1-P03-1033,bq responses </term> to each <term> user </term> in <term> spoken dialogue systems </term> . Unlike previous studies that focus
other,6-2-P03-1033,bq Unlike previous studies that focus on <term> user </term> 's <term> knowledge </term> or typical
other,8-2-P03-1033,bq studies that focus on <term> user </term> 's <term> knowledge </term> or typical kinds of <term> users </term>
other,13-2-P03-1033,bq knowledge </term> or typical kinds of <term> users </term> , the <term> user model </term> we propose
model,16-2-P03-1033,bq typical kinds of <term> users </term> , the <term> user model </term> we propose is more comprehensive
model,8-3-P03-1033,bq Specifically , we set up three dimensions of <term> user models </term> : <term> skill level </term> to the <term>
other,11-3-P03-1033,bq dimensions of <term> user models </term> : <term> skill level </term> to the <term> system </term> , <term> knowledge
tech,15-3-P03-1033,bq </term> : <term> skill level </term> to the <term> system </term> , <term> knowledge level </term> on the
other,17-3-P03-1033,bq level </term> to the <term> system </term> , <term> knowledge level </term> on the <term> target domain </term> and
other,21-3-P03-1033,bq <term> knowledge level </term> on the <term> target domain </term> and the degree of <term> hastiness </term>
other,27-3-P03-1033,bq target domain </term> and the degree of <term> hastiness </term> . Moreover , the <term> models </term>
model,3-4-P03-1033,bq <term> hastiness </term> . Moreover , the <term> models </term> are automatically derived by <term>
tech,8-4-P03-1033,bq </term> are automatically derived by <term> decision tree learning </term> using real <term> dialogue data </term>
lr,13-4-P03-1033,bq decision tree learning </term> using real <term> dialogue data </term> collected by the <term> system </term>
tech,18-4-P03-1033,bq dialogue data </term> collected by the <term> system </term> . We obtained reasonable <term> classification
measure(ment),3-5-P03-1033,bq system </term> . We obtained reasonable <term> classification accuracy </term> for all dimensions . <term> Dialogue
other,0-6-P03-1033,bq accuracy </term> for all dimensions . <term> Dialogue strategies </term> based on the <term> user modeling </term>
skrij podrobnosti