tech,9-1-H01-1049,bq paradigm for <term> human interaction with data sources </term> . We integrate a <term> spoken
lr,8-6-N01-1003,bq automatically learned from <term> training data </term> . We show that the trained <term> SPR
lr,19-2-P01-1004,bq both <term> character - and word-segmented data </term> , in combination with a range of <term>
other,34-2-P01-1047,bq learning algorithm </term> from <term> structured data </term> ( based on a <term> typing-algorithm
lr,15-1-N03-1001,bq manual transcription </term> of <term> training data </term> . The method combines <term> domain
lr,7-4-N03-1012,bq <term> system </term> against the <term> annotated data </term> shows that , it successfully classifies
lr,2-1-N03-2003,bq </term> result . Sources of <term> training data </term> suitable for <term> language modeling
lr,7-2-N03-2003,bq In this paper , we show how <term> training data </term> can be supplemented with <term> text
lr,43-2-N03-2003,bq bigger performance gains from the <term> data </term> by using <term> class-dependent interpolation
other,27-1-N03-4004,bq inflow of <term> multilingual , multimedia data </term> . It gives users the ability to spend
ability to spend their time finding more data relevant to their task , and gives them
other,15-3-N03-4010,bq browsing the <term> repository </term> of <term> data objects </term> created by the <term> system
other,13-1-P03-1005,bq </term> for <term> structured natural language data </term> . The <term> HDAG Kernel </term> directly
other,15-1-P03-1009,bq classes </term> from undisambiguated <term> corpus data </term> . We describe a new approach which
other,30-4-P03-1009,bq classifying </term><term> undisambiguated SCF data </term> . We apply a <term> decision tree based
lr,13-4-P03-1033,bq learning </term> using real <term> dialogue data </term> collected by the <term> system </term>
lr,14-1-P03-1058,bq is the lack of <term> manually sense-tagged data </term> required for <term> supervised learning
lr,11-2-P03-1058,bq automatically acquire <term> sense-tagged training data </term> from <term> English-Chinese parallel
tech,5-3-P03-1058,bq this <term> method of acquiring sense-tagged data </term> is promising . On a subset of the
lr,37-4-P03-1058,bq advantage that <term> manually sense-tagged data </term> have in their <term> sense coverage
hide detail