From Frequency to Meaning: Vector Space Models of Semantics

TL;DR

This paper surveys vector space models of semantics, highlighting their structures, applications, and open source projects to demonstrate their role in advancing computational understanding of human language meaning.

Contribution

It provides a comprehensive overview of VSMs, categorizing them by matrix structure and applications, and offers insights into their practical and theoretical significance.

Findings

Three main classes of VSMs identified: term-document, word-context, pair-pattern.

VSMs are applied across diverse semantic processing tasks.

Open source projects exemplify practical implementations of VSMs.

Abstract

Computers understand very little of the meaning of human language. This profoundly limits our ability to give instructions to computers, the ability of computers to explain their actions to us, and the ability of computers to analyse and process text. Vector space models (VSMs) of semantics are beginning to address these limits. This paper surveys the use of VSMs for semantic processing of text. We organize the literature on VSMs according to the structure of the matrix in a VSM. There are currently three broad classes of VSMs, based on term-document, word-context, and pair-pattern matrices, yielding three classes of applications. We survey a broad range of applications in these three categories and we take a detailed look at a specific open source project in each category. Our goal in this survey is to show the breadth of applications of VSMs for semantics, to provide a new perspective on VSMs for those who are already familiar with the area, and to provide pointers into the literature for those who are less familiar with the field.