As grapheme­to­phoneme methods proliferate, their careful evaluation becomes increasingly important. This paper ex­ plores a variety of metrics to compare the automatic pronunci­ ation methods of three freely­available grapheme­to­phoneme packages on a large dictionary. Two metrics, presented here for the first time, rely upon a novel weighted phonemic substi­ tution matrix ...

BACKGROUND main biomedical information retrieval systems are based on controlled vocabularies and most specifically on terminologies or ontologies (T/O). These classification structures allow indexing, coding, annotating different kind of documents. Many T/O have been created for different purposes and it became a problem for finding specific concepts in the multitude of existing nomenclatures....

We use the grammatical relations (GRs) described in Carroll et al. (1998) to compare a number of parsing algorithms A first ranking of the parsers is provided by comparing the extracted GRs to a gold standard GR annotation of 500 Susanne sentences: this required an implementation of GR extraction software for Penn Treebank style parsers. In addition, we perform an experiment using the extracted...

Recently, a new technique called compare-by-hash has become popular. Compare-by-hash is a method of content-based addressing in which data is identified only by the cryptographic hash of its contents. Hash collisions are ignored, with the justification that they occur less often than many kinds of hardware errors. Compare-by-hash is a powerful, versatile tool in the software architect’s bag of ...