In this paper we report the results of four experiments conducted to extract lists of nouns that exhibit inherent polysemy from corpus data following semiautomatic and automatic procedures. We compare the methods used ...