Semantic spaces: measuring the distance between different subspaces

Guido Zuccon, Leif A. Azzopardi, C. J. Rijsbergen

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

22 Citations (Scopus)

Abstract

Semantic Space models, which provide a numerical representation of words’ meaning extracted from corpus of documents, have been formalized in terms of Hermitian operators over real valued Hilbert spaces by Bruza et al. [1]. The collapse of a word into a particular meaning has been investigated applying the notion of quantum collapse of superpositional states [2]. While the semantic association between words in a Semantic Space can be computed by means of the Minkowski distance [3] or the cosine of the angle between the vector representation of each pair of words, a new procedure is needed in order to establish relations between two or more Semantic Spaces. We address the question: how can the distance between different Semantic Spaces be computed? By representing each Semantic Space as a subspace of a more general Hilbert space, the relationship between Semantic Spaces can be computed by means of the subspace distance. Such distance needs to take into account the difference in the dimensions between subspaces. The availability of a distance for comparing different Semantic Subspaces would enable to achieve a deeper understanding about the geometry of Semantic Spaces which would possibly translate into better effectiveness in Information Retrieval tasks.
LanguageEnglish
Title of host publicationQuantum Interaction
Subtitle of host publicationThird International Symposium, QI 2009, Saarbrücken, Germany, March 25-27, 2009. Proceedings
EditorsPeter Bruza, Donald Sofge, William Lawless, Keith van Rijsbergen, Matthias Klusch
Place of PublicationBerlin, Heidelberg
PublisherSpringer-Verlag
Pages225-236
Number of pages12
ISBN (Print)978-3-642-00833-7
DOIs
Publication statusPublished - 9 Mar 2009
Externally publishedYes

Publication series

NameLecture Notes in Computer Science
PublisherSpringer Verlag
Volume5494

Fingerprint

semantics
information retrieval
mathematics

Keywords

  • semantic space models
  • information retrieval
  • semantic subspaces

Cite this

Zuccon, G., Azzopardi, L. A., & Rijsbergen, C. J. (2009). Semantic spaces: measuring the distance between different subspaces. In P. Bruza, D. Sofge, W. Lawless, K. van Rijsbergen, & M. Klusch (Eds.), Quantum Interaction: Third International Symposium, QI 2009, Saarbrücken, Germany, March 25-27, 2009. Proceedings (pp. 225-236). (Lecture Notes in Computer Science; Vol. 5494). Berlin, Heidelberg: Springer-Verlag. https://doi.org/10.1007/978-3-642-00834-4_19
Zuccon, Guido ; Azzopardi, Leif A. ; Rijsbergen, C. J. / Semantic spaces : measuring the distance between different subspaces. Quantum Interaction: Third International Symposium, QI 2009, Saarbrücken, Germany, March 25-27, 2009. Proceedings. editor / Peter Bruza ; Donald Sofge ; William Lawless ; Keith van Rijsbergen ; Matthias Klusch. Berlin, Heidelberg : Springer-Verlag, 2009. pp. 225-236 (Lecture Notes in Computer Science).
@inproceedings{f4b3d77383404c629719701c1f0f58e0,
title = "Semantic spaces: measuring the distance between different subspaces",
abstract = "Semantic Space models, which provide a numerical representation of words’ meaning extracted from corpus of documents, have been formalized in terms of Hermitian operators over real valued Hilbert spaces by Bruza et al. [1]. The collapse of a word into a particular meaning has been investigated applying the notion of quantum collapse of superpositional states [2]. While the semantic association between words in a Semantic Space can be computed by means of the Minkowski distance [3] or the cosine of the angle between the vector representation of each pair of words, a new procedure is needed in order to establish relations between two or more Semantic Spaces. We address the question: how can the distance between different Semantic Spaces be computed? By representing each Semantic Space as a subspace of a more general Hilbert space, the relationship between Semantic Spaces can be computed by means of the subspace distance. Such distance needs to take into account the difference in the dimensions between subspaces. The availability of a distance for comparing different Semantic Subspaces would enable to achieve a deeper understanding about the geometry of Semantic Spaces which would possibly translate into better effectiveness in Information Retrieval tasks.",
keywords = "semantic space models, information retrieval, semantic subspaces",
author = "Guido Zuccon and Azzopardi, {Leif A.} and Rijsbergen, {C. J.}",
year = "2009",
month = "3",
day = "9",
doi = "10.1007/978-3-642-00834-4_19",
language = "English",
isbn = "978-3-642-00833-7",
series = "Lecture Notes in Computer Science",
publisher = "Springer-Verlag",
pages = "225--236",
editor = "Peter Bruza and Donald Sofge and William Lawless and {van Rijsbergen}, Keith and Matthias Klusch",
booktitle = "Quantum Interaction",

}

Zuccon, G, Azzopardi, LA & Rijsbergen, CJ 2009, Semantic spaces: measuring the distance between different subspaces. in P Bruza, D Sofge, W Lawless, K van Rijsbergen & M Klusch (eds), Quantum Interaction: Third International Symposium, QI 2009, Saarbrücken, Germany, March 25-27, 2009. Proceedings. Lecture Notes in Computer Science, vol. 5494, Springer-Verlag, Berlin, Heidelberg, pp. 225-236. https://doi.org/10.1007/978-3-642-00834-4_19

Semantic spaces : measuring the distance between different subspaces. / Zuccon, Guido; Azzopardi, Leif A.; Rijsbergen, C. J.

Quantum Interaction: Third International Symposium, QI 2009, Saarbrücken, Germany, March 25-27, 2009. Proceedings. ed. / Peter Bruza; Donald Sofge; William Lawless; Keith van Rijsbergen; Matthias Klusch. Berlin, Heidelberg : Springer-Verlag, 2009. p. 225-236 (Lecture Notes in Computer Science; Vol. 5494).

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

TY - GEN

T1 - Semantic spaces

T2 - measuring the distance between different subspaces

AU - Zuccon, Guido

AU - Azzopardi, Leif A.

AU - Rijsbergen, C. J.

PY - 2009/3/9

Y1 - 2009/3/9

N2 - Semantic Space models, which provide a numerical representation of words’ meaning extracted from corpus of documents, have been formalized in terms of Hermitian operators over real valued Hilbert spaces by Bruza et al. [1]. The collapse of a word into a particular meaning has been investigated applying the notion of quantum collapse of superpositional states [2]. While the semantic association between words in a Semantic Space can be computed by means of the Minkowski distance [3] or the cosine of the angle between the vector representation of each pair of words, a new procedure is needed in order to establish relations between two or more Semantic Spaces. We address the question: how can the distance between different Semantic Spaces be computed? By representing each Semantic Space as a subspace of a more general Hilbert space, the relationship between Semantic Spaces can be computed by means of the subspace distance. Such distance needs to take into account the difference in the dimensions between subspaces. The availability of a distance for comparing different Semantic Subspaces would enable to achieve a deeper understanding about the geometry of Semantic Spaces which would possibly translate into better effectiveness in Information Retrieval tasks.

AB - Semantic Space models, which provide a numerical representation of words’ meaning extracted from corpus of documents, have been formalized in terms of Hermitian operators over real valued Hilbert spaces by Bruza et al. [1]. The collapse of a word into a particular meaning has been investigated applying the notion of quantum collapse of superpositional states [2]. While the semantic association between words in a Semantic Space can be computed by means of the Minkowski distance [3] or the cosine of the angle between the vector representation of each pair of words, a new procedure is needed in order to establish relations between two or more Semantic Spaces. We address the question: how can the distance between different Semantic Spaces be computed? By representing each Semantic Space as a subspace of a more general Hilbert space, the relationship between Semantic Spaces can be computed by means of the subspace distance. Such distance needs to take into account the difference in the dimensions between subspaces. The availability of a distance for comparing different Semantic Subspaces would enable to achieve a deeper understanding about the geometry of Semantic Spaces which would possibly translate into better effectiveness in Information Retrieval tasks.

KW - semantic space models

KW - information retrieval

KW - semantic subspaces

U2 - 10.1007/978-3-642-00834-4_19

DO - 10.1007/978-3-642-00834-4_19

M3 - Conference contribution book

SN - 978-3-642-00833-7

T3 - Lecture Notes in Computer Science

SP - 225

EP - 236

BT - Quantum Interaction

A2 - Bruza, Peter

A2 - Sofge, Donald

A2 - Lawless, William

A2 - van Rijsbergen, Keith

A2 - Klusch, Matthias

PB - Springer-Verlag

CY - Berlin, Heidelberg

ER -

Zuccon G, Azzopardi LA, Rijsbergen CJ. Semantic spaces: measuring the distance between different subspaces. In Bruza P, Sofge D, Lawless W, van Rijsbergen K, Klusch M, editors, Quantum Interaction: Third International Symposium, QI 2009, Saarbrücken, Germany, March 25-27, 2009. Proceedings. Berlin, Heidelberg: Springer-Verlag. 2009. p. 225-236. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-642-00834-4_19