Investigation into Human Preference between Common and Unambiguous Lexical Substitutions

We present a study that investigates the fac tors that determine what makes a good lexi cal substitution. We begin by observing that there is a correlation between the corpus fre quency of words and the number of WordNet senses they have, and hypothesise that readers might prefer common, but ambiguous terms over less ambiguous but less common words. In this paper we identify four properties of a word that determine whether it is a suit able substitution in a given context, and ask volunteers to rank their preferences between two common but ambiguous lexical substi tutions, and two uncommon but also unambiguous ones. Preliminary results suggest a slight preference towards the unambiguous.