support linking of <term> lexical entries </term> to their <term> translation equivalents </term>
their <term> translation equivalents </term> . To help this task we have developed an <term>
</term> . Several experiments , corresponding to rather <term> closed semantic domains </term>
domains </term> , have been developed in order to generate <term> lexical cross-relations </term>
<term> inheritance mechanisms </term> are used to capture <term> lexical generalizations </term>
</term> in a program roughly corresponding to <term> arcs </term> in a <term> chart </term> .
the <term> parsing complexity </term> reduces to that of standard <term> chart-based algorithms
<term> extraction and conversion rules </term> to <term> entries </term> from <term> machine readable
strategies </term> , in a similar fashion to the <term> SYSCONJ system </term> developed
<term> language model </term> that attempts to capture <term> long distance constraints </term>
mixture trigram models </term> as compared to using a <term> trigram model </term> . This
<term> speechrepairs </term> , and so is able to mark a <term> transition </term> either as
hide detail