tech,6-9-J05-1003,bq The article also introduces a new <term> algorithm </term> for the <term> boosting approach </term>
tech,8-10-J05-1003,bq significant efficiency gains for the new <term> algorithm </term> over the obvious <term> implementation
tech,4-4-J05-1003,bq </term> as evidence . The strength of our <term> approach </term> is that it allows a <term> tree </term>
tech,16-12-J05-1003,bq language parsing ( NLP ) </term> , the <term> approach </term> should be applicable to many other
model,7-7-J05-1003,bq <term> log-likelihood </term> under a <term> baseline model </term> ( that of <term> Collins [ 1999 ] </term>
measure(ment),18-8-J05-1003,bq <term> F-measure </term> error over the <term> baseline model ’s score </term> of 88.2 % . The article also introduces
tech,13-5-J05-1003,bq reranking task </term> , based on the <term> boosting approach </term> to <term> ranking problems </term> described
tech,9-9-J05-1003,bq a new <term> algorithm </term> for the <term> boosting approach </term> which takes advantage of the <term>
tech,15-10-J05-1003,bq obvious <term> implementation </term> of the <term> boosting approach </term> . We argue that the method is an
tech,3-6-J05-1003,bq al. ( 1998 ) </term> . We apply the <term> boosting method </term> to <term> parsing </term> the <term> Wall
other,7-2-J05-1003,bq <term> parser </term> produces a set of <term> candidate parses </term> for each input <term> sentence </term>
other,12-7-J05-1003,bq <term> baseline model </term> ( that of <term> Collins [ 1999 ] </term> ) with evidence from an additional
other,37-4-J05-1003,bq overlap and without the need to define a <term> derivation </term> or a <term> generative model </term>
tech,21-11-J05-1003,bq simplicity and efficiency — to work on <term> feature selection methods </term> within <term> log-linear ( maximum-entropy
other,14-3-J05-1003,bq <term> ranking </term> , using additional <term> features </term> of the <term> tree </term> as evidence
other,19-4-J05-1003,bq represented as an arbitrary set of <term> features </term> , without concerns about how these
other,26-4-J05-1003,bq , without concerns about how these <term> features </term> interact or overlap and without the
other,45-4-J05-1003,bq generative model </term> which takes these <term> features </term> into account . We introduce a new
other,23-7-J05-1003,bq evidence from an additional 500,000 <term> features </term> over <term> parse trees </term> that
measure(ment),6-8-J05-1003,bq <term> model </term> achieved 89.75 % <term> F-measure </term> , a 13 % relative decrease in <term>
hide detail