TDM modeling and evaluation of different domain transforms for LSI

Tareq Jaber, Abbes Amira, Peter Milligan

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Latent semantic indexing (LSI) is a popular technique used in information retrieval (IR) applications. This paper presents a novel evaluation strategy based on the use of image processing tools. The authors evaluate the use of the discrete cosine transform (DCT) and Cohen Daubechies Feauveau 9/7 (CDF9/7) wavelet transform as a preprocessing step for the singular value decomposition (SVD) step of the LSI system. In addition, the effect of different threshold types on the search results is examined. The results show that accuracy can be increased by applying both transforms as a preprocessing step, with better performance for the hard-threshold function. The choice of the best threshold value is a key factor in the transform process. This paper also describes the most effective structure for the database to facilitate efficient searching in the LSI system.
Original languageEnglish
Pages (from-to)2406-2417
Number of pages12
JournalNeurocomputing
Volume72
Issue number10-12
DOIs
Publication statusPublished - Jun 2009
Externally publishedYes

Fingerprint

Time division multiplexing
Semantics
Wavelet Analysis
Discrete cosine transforms
Information Storage and Retrieval
Singular value decomposition
Information retrieval
Wavelet transforms
Image processing
Databases

Keywords

  • Latent semantic indexing
  • Information retrieval
  • Discrete cosine transform
  • Singular value decomposition
  • Cohen Daubechies Feauveau 9/7
  • Hard thresholding
  • Soft thresholding

Cite this

Jaber, Tareq ; Amira, Abbes ; Milligan, Peter. / TDM modeling and evaluation of different domain transforms for LSI. In: Neurocomputing. 2009 ; Vol. 72, No. 10-12. pp. 2406-2417.
@article{766c47a3f4714a26af474e0bb1f9fc0f,
title = "TDM modeling and evaluation of different domain transforms for LSI",
abstract = "Latent semantic indexing (LSI) is a popular technique used in information retrieval (IR) applications. This paper presents a novel evaluation strategy based on the use of image processing tools. The authors evaluate the use of the discrete cosine transform (DCT) and Cohen Daubechies Feauveau 9/7 (CDF9/7) wavelet transform as a preprocessing step for the singular value decomposition (SVD) step of the LSI system. In addition, the effect of different threshold types on the search results is examined. The results show that accuracy can be increased by applying both transforms as a preprocessing step, with better performance for the hard-threshold function. The choice of the best threshold value is a key factor in the transform process. This paper also describes the most effective structure for the database to facilitate efficient searching in the LSI system.",
keywords = "Latent semantic indexing, Information retrieval, Discrete cosine transform, Singular value decomposition, Cohen Daubechies Feauveau 9/7, Hard thresholding, Soft thresholding",
author = "Tareq Jaber and Abbes Amira and Peter Milligan",
year = "2009",
month = "6",
doi = "10.1016/j.neucom.2008.12.010",
language = "English",
volume = "72",
pages = "2406--2417",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier B.V.",
number = "10-12",

}

TDM modeling and evaluation of different domain transforms for LSI. / Jaber, Tareq; Amira, Abbes; Milligan, Peter.

In: Neurocomputing, Vol. 72, No. 10-12, 06.2009, p. 2406-2417.

Research output: Contribution to journalArticle

TY - JOUR

T1 - TDM modeling and evaluation of different domain transforms for LSI

AU - Jaber, Tareq

AU - Amira, Abbes

AU - Milligan, Peter

PY - 2009/6

Y1 - 2009/6

N2 - Latent semantic indexing (LSI) is a popular technique used in information retrieval (IR) applications. This paper presents a novel evaluation strategy based on the use of image processing tools. The authors evaluate the use of the discrete cosine transform (DCT) and Cohen Daubechies Feauveau 9/7 (CDF9/7) wavelet transform as a preprocessing step for the singular value decomposition (SVD) step of the LSI system. In addition, the effect of different threshold types on the search results is examined. The results show that accuracy can be increased by applying both transforms as a preprocessing step, with better performance for the hard-threshold function. The choice of the best threshold value is a key factor in the transform process. This paper also describes the most effective structure for the database to facilitate efficient searching in the LSI system.

AB - Latent semantic indexing (LSI) is a popular technique used in information retrieval (IR) applications. This paper presents a novel evaluation strategy based on the use of image processing tools. The authors evaluate the use of the discrete cosine transform (DCT) and Cohen Daubechies Feauveau 9/7 (CDF9/7) wavelet transform as a preprocessing step for the singular value decomposition (SVD) step of the LSI system. In addition, the effect of different threshold types on the search results is examined. The results show that accuracy can be increased by applying both transforms as a preprocessing step, with better performance for the hard-threshold function. The choice of the best threshold value is a key factor in the transform process. This paper also describes the most effective structure for the database to facilitate efficient searching in the LSI system.

KW - Latent semantic indexing

KW - Information retrieval

KW - Discrete cosine transform

KW - Singular value decomposition

KW - Cohen Daubechies Feauveau 9/7

KW - Hard thresholding

KW - Soft thresholding

U2 - 10.1016/j.neucom.2008.12.010

DO - 10.1016/j.neucom.2008.12.010

M3 - Article

VL - 72

SP - 2406

EP - 2417

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

IS - 10-12

ER -