A Compositional Distributional Semantics, Two Concrete Constructions, and some Experimental Evaluations

TL;DR

This paper reviews a hybrid compositional distributional semantics model that uses categorical methods to derive sentence meanings from word vectors, demonstrating its application with toy and real data, including disambiguation tasks.

Contribution

It introduces concrete constructions for the model and techniques to build meaning vectors, bridging theoretical framework with practical implementations.

Findings

Effective vector space constructions for words and sentences

Successful application to real corpus data

Disambiguation experiments show promising results

Abstract

We provide an overview of the hybrid compositional distributional model of meaning, developed in Coecke et al. (arXiv:1003.4394v1 [cs.CL]), which is based on the categorical methods also applied to the analysis of information flow in quantum protocols. The mathematical setting stipulates that the meaning of a sentence is a linear function of the tensor products of the meanings of its words. We provide concrete constructions for this definition and present techniques to build vector spaces for meaning vectors of words, as well as that of sentences. The applicability of these methods is demonstrated via a toy vector space as well as real data from the British National Corpus and two disambiguation experiments.