tech,9-1-P06-2001,ak paper , we describe the research using <term> machine learning techniques </term> to build a <term> comma checker </term>
tech,15-1-P06-2001,ak learning techniques </term> to build a <term> comma checker </term> to be integrated in a <term> grammar
tech,22-1-P06-2001,ak checker </term> to be integrated in a <term> grammar checker </term> for <term> Basque </term> . After several
other,25-1-P06-2001,ak in a <term> grammar checker </term> for <term> Basque </term> . After several experiments , and
lr,9-2-P06-2001,ak experiments , and trained with a little <term> corpus </term> of 100,000 words , the system guesses
other,20-2-P06-2001,ak system guesses correctly not placing <term> commas </term> with a <term> precision </term> of 96
measure(ment),23-2-P06-2001,ak not placing <term> commas </term> with a <term> precision </term> of 96 % and a <term> recall </term> of
measure(ment),29-2-P06-2001,ak <term> precision </term> of 96 % and a <term> recall </term> of 98 % . It also gets a <term> precision
measure(ment),4-3-P06-2001,ak recall </term> of 98 % . It also gets a <term> precision </term> of 70 % and a <term> recall </term> of
measure(ment),10-3-P06-2001,ak <term> precision </term> of 70 % and a <term> recall </term> of 49 % in the task of placing <term>
other,19-3-P06-2001,ak </term> of 49 % in the task of placing <term> commas </term> . Finally , we have shown that these
lr,18-4-P06-2001,ak using a bigger and a more homogeneous <term> corpus </term> to train , that is , a bigger <term>
lr,27-4-P06-2001,ak </term> to train , that is , a bigger <term> corpus </term> written by one unique author . This
skrij podrobnosti