Clustered SVD strategies in latent semantic indexing

Jing Gao; Jun Zhang

doi:10.1016/j.ipm.2004.10.005

Clustered SVD strategies in latent semantic indexing

Jing Gao, Jun Zhang

Computer Science

Research output: Contribution to journal › Article › peer-review

54 Scopus citations

Abstract

The text retrieval method using latent semantic indexing (LSI) technique with truncated singular value decomposition (SVD) has been intensively studied in recent years. The SVD reduces the noise contained in the original representation of the term-document matrix and improves the information retrieval accuracy. Recent studies indicate that SVD is mostly useful for small homogeneous data collections. For large inhomogeneous datasets, the performance of the SVD based text retrieval technique may deteriorate. We propose to partition a large inhomogeneous dataset into several smaller ones with clustered structure, on which we apply the truncated SVD. Our experimental results show that the clustered SVD strategies may enhance the retrieval accuracy and reduce the computing and storage costs.

Original language	English
Pages (from-to)	1051-1063
Number of pages	13
Journal	Information Processing and Management
Volume	41
Issue number	5
DOIs	https://doi.org/10.1016/j.ipm.2004.10.005
State	Published - Sep 2005

Bibliographical note

Funding Information:
The research work of the authors was supported in part by the US National Science Foundation under grants CCR-9988165, CCR-0092532, ACR-0202934, and ACR-0234270, and by the US Department of Energy Office of Science under grant DE-FG02-02ER45961.

Keywords

Clustering
Latent semantic indexing
SVD
Text retrieval

ASJC Scopus subject areas

Information Systems
Media Technology
Computer Science Applications
Management Science and Operations Research
Library and Information Sciences

Access to Document

10.1016/j.ipm.2004.10.005

Cite this

@article{1b5718e03b784abdbe6883483ae4d534,

title = "Clustered SVD strategies in latent semantic indexing",

abstract = "The text retrieval method using latent semantic indexing (LSI) technique with truncated singular value decomposition (SVD) has been intensively studied in recent years. The SVD reduces the noise contained in the original representation of the term-document matrix and improves the information retrieval accuracy. Recent studies indicate that SVD is mostly useful for small homogeneous data collections. For large inhomogeneous datasets, the performance of the SVD based text retrieval technique may deteriorate. We propose to partition a large inhomogeneous dataset into several smaller ones with clustered structure, on which we apply the truncated SVD. Our experimental results show that the clustered SVD strategies may enhance the retrieval accuracy and reduce the computing and storage costs.",

keywords = "Clustering, Latent semantic indexing, SVD, Text retrieval",

author = "Jing Gao and Jun Zhang",

note = "Funding Information: The research work of the authors was supported in part by the US National Science Foundation under grants CCR-9988165, CCR-0092532, ACR-0202934, and ACR-0234270, and by the US Department of Energy Office of Science under grant DE-FG02-02ER45961. ",

year = "2005",

month = sep,

doi = "10.1016/j.ipm.2004.10.005",

language = "English",

volume = "41",

pages = "1051--1063",

number = "5",

}

TY - JOUR

T1 - Clustered SVD strategies in latent semantic indexing

AU - Gao, Jing

AU - Zhang, Jun

N1 - Funding Information: The research work of the authors was supported in part by the US National Science Foundation under grants CCR-9988165, CCR-0092532, ACR-0202934, and ACR-0234270, and by the US Department of Energy Office of Science under grant DE-FG02-02ER45961.

PY - 2005/9

Y1 - 2005/9

N2 - The text retrieval method using latent semantic indexing (LSI) technique with truncated singular value decomposition (SVD) has been intensively studied in recent years. The SVD reduces the noise contained in the original representation of the term-document matrix and improves the information retrieval accuracy. Recent studies indicate that SVD is mostly useful for small homogeneous data collections. For large inhomogeneous datasets, the performance of the SVD based text retrieval technique may deteriorate. We propose to partition a large inhomogeneous dataset into several smaller ones with clustered structure, on which we apply the truncated SVD. Our experimental results show that the clustered SVD strategies may enhance the retrieval accuracy and reduce the computing and storage costs.

AB - The text retrieval method using latent semantic indexing (LSI) technique with truncated singular value decomposition (SVD) has been intensively studied in recent years. The SVD reduces the noise contained in the original representation of the term-document matrix and improves the information retrieval accuracy. Recent studies indicate that SVD is mostly useful for small homogeneous data collections. For large inhomogeneous datasets, the performance of the SVD based text retrieval technique may deteriorate. We propose to partition a large inhomogeneous dataset into several smaller ones with clustered structure, on which we apply the truncated SVD. Our experimental results show that the clustered SVD strategies may enhance the retrieval accuracy and reduce the computing and storage costs.

KW - Clustering

KW - Latent semantic indexing

KW - SVD

KW - Text retrieval

UR - http://www.scopus.com/inward/record.url?scp=16244363649&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=16244363649&partnerID=8YFLogxK

U2 - 10.1016/j.ipm.2004.10.005

DO - 10.1016/j.ipm.2004.10.005

M3 - Article

AN - SCOPUS:16244363649

SN - 0306-4573

VL - 41

SP - 1051

EP - 1063

JO - Information Processing and Management

JF - Information Processing and Management

IS - 5

ER -

Clustered SVD strategies in latent semantic indexing

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this