Paradigmatic Modifiability Statistics for the Extraction of Complex Multi-Word Terms

We here propose a new method which sets apart domain-specific terminology from common non-specific noun phrases. It is based on the observation that terminological multi-word groups reveal a considerably lesser degree of distributional variation than non-specific noun phrases. We define a measure for the observable amount of paradigmatic modifiability of terms and, subsequently, test it on bigram, trigram and quadgram noun phrases extracted from a 104-million-word biomedical text corpus. Using a community-wide curated biomedical terminology system as an evaluation gold standard, we show that our algorithm significantly outperforms a variety of standard term identification measures. We also provide empirical evidence that our methodolgy is essentially domain- and corpus-size-independent.