The question of how to compose meaning in distributional representations of meaning has recently been recognised as a central issue in computational linguistics. In this paper we describe three general and powerful tools that can be used to describe composition in distributional semantics: quotient algebras, learning of finite dimensional algebras, and the construction of algebras from semigroups.

In this article, we explore an integration of a formal semantic approach to lexical meaning and an approach based on distributional methods. First, we outline a formal semantic theory that aims to combine the virtues of both formal and distributional frameworks. We then proceed to develop an algebraic interpretation of that formal semantic theory and show how at least two kinds of distributiona...

Approaches to determining the factuality of diagnoses and findings in clinical text tend to rely on dictionaries of marker words for uncertainty and negation. Here, a method for semi-automatically expanding a dictionary of marker words using distributional semantics is presented and evaluated. It is shown that ranking candidates for inclusion according to their proximity to cluster centroids of...

We introduce two ways to detect entailment using distributional semantic representations of phrases. Our first experiment shows that the entailment relation between adjective-noun constructions and their head nouns (big cat |= cat), once represented as semantic vector pairs, generalizes to lexical entailment among nouns (dog |= animal). Our second experiment shows that a classifier fed semantic...

The paper presents a procedure of building an evaluation dataset1. for the validation of compositional distributional semantics models estimated for languages other than English. The procedure generally builds on steps designed to assemble the SICK corpus, which contains pairs of English sentences annotated for semantic relatedness and entailment, because we aim at building a comparable dataset...

Previous approaches to the problem of measuring similarity between automatically generated topics have been based on comparison of the topics’ word probability distributions. This paper presents alternative approaches, including ones based on distributional semantics and knowledgebased measures, evaluated by comparison with human judgements. The best performing methods provide reliable estimate...

Distributional representations have recently been proposed as a general-purpose representation of natural language meaning, to replace logical form. There is, however, one important difference between logical and distributional representations: Logical languages have a clear semantics, while distributional representations do not. In this paper, we propose a semantics for distributional represen...

Distributional learning studies investigating the acquisition of the Dutch contrast /ɑ/-/a:/ by nonnative Dutch learners have reported mixed results. The present study extends the literature by examining whether (i) naïve listeners are able to extract the distribution structure of a sequence of/ɑ//aː/ tokens drawn from a continuum,; and (ii) differential effects exist between naturalistic vs. e...

Since synonyms are important lexical knowledge, various methods have been proposed for automatic synonym acquisition. Whereas most of the methods are based on the distributional hypothesis and utilize contextual clues, little attention has been paid to what kind of contextual information is useful for the purpose. As one of the ways to augment contextual information, we propose the use of indir...

We document the relationship between distributional preferences and voting decisions in a large and diverse sample of Americans. Using a generalized dictator game, we generate individual-level measures of fair-mindedness (weight on oneself versus others) and equality-efficiency tradeoffs. Subjects' equality-efficiency tradeoffs predict their political decisions: equality-focused subjects are mo...