Exploitation of Co-reference in Distributional Semantics

Exploitation of Co-reference in Distributional Semantics Dominik Schlechtweg University of Stuttgart Stuttgart, Germany dominik.schlechtweg@ling.uni-stuttgart.de Abstract The aim of distributional semantics is to model the similarity of the meaning of words via the words they occur with. Thereby, it relies on the distributional hypothesis implying that similar words have similar contexts. Deducing meaning from the distribution of words is interesting as it can be done automatically on large amounts of freely available raw text. It is because of this convenience that most current state-of-the-art-models of distributional semantics operate on raw text, although there have been successful attempts to integrate other kinds of—e.g., syntactic—information to improve distributional semantic models. In contrast, less attention has been paid to semantic information in the research community. One reason for this is that the extraction of semantic information from raw text is a complex, elaborate matter and in great parts not yet satisfyingly solved. Recently, however, there have been successful attempts to integrate a certain kind of semantic information, i.e., co-reference. Two basically different kinds of information contributed by co-reference with respect to the distribution of words will be identified. We will then focus on one of these and examine its general potential to improve distributional semantic models as well as certain more specific hypotheses. Keywords: Distributional Semantics, Co-reference, Vector Space Models 1. Introduction The original aim of distributional semantics is to model the similarity of the meaning—the semantics—of words. The basic assumption underlying this approach is that the se- mantic similarity of two words is a function of their con- texts. That is, in other words, the meaning of words can be inferred from the frequencies of the words they imme- diately occur with and this can happen in such a way that the degree of similarity of those meanings can be measured. Semantic similarity is a key concept in the modeling of lan- guage and thus in computational linguistics. It is crucial in a variety of linguistic applications influencing our everyday life such as search engines. In distributional semantics we represent the meaning of a word by a vector. This vector is an abstraction over the contexts in which we find the particular word in question. Here lies the crux of the matter: known algorithms of distri- butional semantics consider only those contexts as relevant to the meaning of a target word which are found as con- texts of the word—the particular combination of letters, the string—in question. However, there are other, particularly definable, contexts which encode some of the meaning of a target word. Consider the following text example: When Cesar took on the case of {{Fella}, {an adorable Jack Russell/Italian Greyhound mix}}, {the little dog}’s antics were about to get {his} owner slapped with an eviction notice. At the apartment complex where {Fella} resided, {he} barked nonstop the entire time {his} adoptive mom was at work, ceasing only once she came home at night.1 Building up a vector representation for the meaning of dog, a standard algorithm of distributional semantics would 1Cesar …

Content truncated to 500 words