<?xml version="1.0" standalone="yes"?>
<Paper uid="N01-1009">
  <Title>A Corpus-based Account of Regular Polysemy: The Case of Context-sensitive Adjectives</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Much recent work in lexical semantics has been concerned with accounting for regular polysemy, i.e., the regular and predictable sense alternations certain classes of words are subject to. Adjectives, more than other categories, are a striking example of regular polysemy since they are able to take on different meanings depending on their context, viz., the noun or noun class they modify (see Pustejovsky (1995) and the references therein).</Paragraph>
    <Paragraph position="1"> The adjective fast in (1) receives different interpretations when modifying the nouns programmer, plane and scientist.Afast programmer is typically a programmer who programs quickly, a fast plane is typically a plane that flies quickly, a fast scientist can be a scientist who publishes papers quickly, who performs experiments quickly, who observes something quickly, who reasons, thinks, or runs quickly. Interestingly, adjectives like fast are ambiguous across and within the nouns they modify. A fast plane is not only a plane that flies quickly, but also a plane that lands, takes off, turns, or travels quickly. Even the more restrictive fast programmer allows more than one interpretation. One can easily think of a context where a fast programmer thinks, runs or talks quickly.</Paragraph>
    <Paragraph position="2">  (1) a. fast programmer b. fast plane c. fast scientist  The work reported in this paper was carried out while the author was at the Division of Informatics, University of Edinburgh. (2) a. easy problem b. difficult language c. good cook d. good soup Adjectives like fast have been extensively studied in the lexical semantics literature and their properties have been known at least since Vendler (1968). The meaning of adjective-noun combinations like those in (1) and (2) are usually paraphrased with a verb modified by the adjective in question or its corresponding adverb. For example, an easy problem is &amp;quot;a problem that is easy to solve&amp;quot; or &amp;quot;a problem that one can solve easily&amp;quot;. In order to account for the meaning of these combinations Vendler (1968, 92) points out that &amp;quot;in most cases not one verb, but a family of verbs is needed&amp;quot;. Vendler further observes that the noun figuring in an adjective-noun combination is usually the subject or object of the paraphrasing verb. Although fast usually triggers a verb-subject interpretation (see (1)), easy and difficult trigger verb-object interpretations (see (2a,b)). An easy problem is usually a problem that is easy to solve, whereas a difficult language is a language that is difficult to learn, speak, or write. Adjectives like good allow either verb-subject or verb-object interpretations: a good cook is a cook who cooks well whereas good soup is soup that tastes good or soup that is good to eat.</Paragraph>
    <Paragraph position="3"> Pustejovsky (1995) avoids enumerating the various senses for adjectives like fast by exploiting the semantics of the nouns they modify. Pustejovsky treats nouns as having a qualia structure as part of their lexical entries, which among other things, specifies possible events associated with the entity. For example, the telic (purpose) role of the qualia structure for problem has a value equivalent to solve. When the adjective easy is combined with problem, it predicates over the telic role of problem and consequently the adjective-noun combination receives the interpretation a problem that is easy to solve. Pustejovsky (1995) does not give an exhaustive list of the telic roles a given noun may have. Furthermore, in cases where more than one interpretations are provided (see Vendler (1968)), no information is given with respect to the likelihood of these interpretations. Out-of context, the number of interpretations for fast scientist is virtually unlimited, yet some interpretations are more likely than others: fast scientist is more likely to be a scientist who performs experiments quickly or who publishes quickly than a scientist who draws or drinks quickly.</Paragraph>
    <Paragraph position="4"> In this paper we focus on polysemous adjective-noun combinations (see (1) and (2)) and attempt to address the following questions: (a) Can the meanings of these adjective-noun combinations be acquired automatically from corpora? (b) Can we constrain the number of interpretations by providing a ranking on the set of possible meanings? (c) Can we determine if an adjective has a preference for a verb-subject or verb-object interpretation? We provide a probabilistic model which combines distributional information about how likely it is for any verb to be modified by the adjective in the adjective-noun combination or its corresponding adverb with information about how likely it is for any verb to take the modified noun as its object or subject. We obtain quantitative information about verb-adjective modification and verb-argument relations from the British National Corpus (BNC), a 100 million word collection of samples of written and spoken language from a wide range of sources designed to represent current British English (Burnard, 1995). We evaluate our results by comparing the model's predictions against human judgments and show that the model's ranking of meanings correlates reliably with human intuitions.</Paragraph>
  </Section>
class="xml-element"></Paper>