From Frequency to Meaning: Vector Space Models of Semantics
От частоты к значению: векторные пространственные модели семантики
2010-02-27
SCID: 54.1/fc5xh7qm
Discuss with AI
pair-pattern matricessemantic processingterm-document matricesvector space modelsword-context matrices
Figures from the paper
Abstract (AI)
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.
Key Findings
1
The paper surveys applications across all three categories and examines one open-source project in detail for each category.
2
The survey organizes semantic VSM research by matrix structure, identifying term-document, word-context, and pair-pattern models as three broad classes.
3
The survey provides a unified perspective on the breadth of VSM applications and directs readers to relevant literature.
4
These three VSM classes correspond to distinct groups of semantic text-processing applications.
5
Vector space models are emerging as a framework for addressing computers’ limited ability to represent and process human-language meaning.
Research Object
Vector space models (VSMs) of semantics for text
Research Subject
the organization, classes, and breadth of semantic text-processing applications of vector space models based on term-document, word-context, and pair-pattern matrices
Publication Details
Publication Date
2010-02-27
Journal
Publisher
ISSN
Cited by
2866
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai1
Cited by8
Distributed Representations of Words and Phrases and their Compositionality2013
Enriching Word Vectors with Subword Information2017
Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank2013
Don't count, predict! A systematic comparison of context-counting vs. context-predicting semantic vectors2014
Multimodal Distributional Semantics2014
Distributional Semantics and Linguistic Theory2019
Composition in Distributional Semantics2013
A distributional assessment of rivalry in word formation2023