measure(ment),13-3-C92-1055,ak <term> parameters </term> to maximize the <term> accuracy rate </term> directly . To make the proposed <term>
measure(ment),1-6-C92-1055,ak has been observed in the test . The <term> accuracy rate </term> of <term> syntactic disambiguation </term>
other,23-2-C92-1055,ak statistical approaches </term> , which resolve <term> ambiguities </term> by indirectly and implicitly using
other,9-2-C92-1055,ak insufficient <term> training data </term> and <term> approximation error </term> introduced by the <term> language model
tech,5-1-C92-1055,ak resolution </term> . In this paper , a <term> discrimination and robustness oriented adaptive learning procedure </term> is proposed to deal with the task
model,14-2-C92-1055,ak approximation error </term> introduced by the <term> language model </term> , traditional <term> statistical approaches
tech,29-2-C92-1055,ak by indirectly and implicitly using <term> maximum likelihood method </term> , fail to achieve high performance
other,9-3-C92-1055,ak remedies these problems by adjusting the <term> parameters </term> to maximize the <term> accuracy rate
other,26-4-C92-1055,ak into consideration by enlarging the <term> separation margin </term> between the correct <term> candidate
tech,18-2-C92-1055,ak language model </term> , traditional <term> statistical approaches </term> , which resolve <term> ambiguities </term>
tech,20-1-C92-1055,ak proposed to deal with the task of <term> syntactic ambiguity resolution </term> . Owing to the problem of insufficient
lr,12-4-C92-1055,ak the possible variations between the <term> training corpus </term> and the real tasks are also taken
lr,6-2-C92-1055,ak Owing to the problem of insufficient <term> training data </term> and <term> approximation error </term>
hide detail