Semantic indexing and searching using a Hopfield net

Hsinchun Chen, Yin Zhang, Andrea L. Houston

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

This paper presents a neural network approach to document semantic indexing. A Hopfield net algorithm was used to simulate human associative memory for concept exploration in the domain of computer science and engineering. INSFEC, a collection of more than 320,000 document abstracts from leading journals, was used as the document testbed. Benchmark tests confirmed that three parameters (maximum number of activated nodes, ε - maximum allowable error, and maximum number of iterations] were useful in positively influencing network convergence behavior without negatively impacting central processing unit performance. Another series of benchmark tests was performed to determine the effectiveness of various filtering techniques in reducing the negative impact of noisy input terms. Preliminary user tests confirmed our expectation that the Hopfield net algorithm is potentially useful as an associative memory technique to improve document recall and precision by solving discrepancies between indexer vocabularies and end-user vocabularies.

Original languageEnglish (US)
Pages (from-to)3-18
Number of pages16
JournalJournal of Information Science
Volume24
Issue number1
StatePublished - 1998

Fingerprint

indexing
Semantics
semantics
Data storage equipment
Testbeds
Computer science
Program processors
vocabulary
Neural networks
computer science
neural network
engineering
performance

ASJC Scopus subject areas

  • Information Systems
  • Library and Information Sciences

Cite this

Semantic indexing and searching using a Hopfield net. / Chen, Hsinchun; Zhang, Yin; Houston, Andrea L.

In: Journal of Information Science, Vol. 24, No. 1, 1998, p. 3-18.

Research output: Contribution to journalArticle

Chen, H, Zhang, Y & Houston, AL 1998, 'Semantic indexing and searching using a Hopfield net', Journal of Information Science, vol. 24, no. 1, pp. 3-18.
Chen, Hsinchun ; Zhang, Yin ; Houston, Andrea L. / Semantic indexing and searching using a Hopfield net. In: Journal of Information Science. 1998 ; Vol. 24, No. 1. pp. 3-18.
@article{a07220e64d7d4a26a61b0b85bdd105ac,
title = "Semantic indexing and searching using a Hopfield net",
abstract = "This paper presents a neural network approach to document semantic indexing. A Hopfield net algorithm was used to simulate human associative memory for concept exploration in the domain of computer science and engineering. INSFEC, a collection of more than 320,000 document abstracts from leading journals, was used as the document testbed. Benchmark tests confirmed that three parameters (maximum number of activated nodes, ε - maximum allowable error, and maximum number of iterations] were useful in positively influencing network convergence behavior without negatively impacting central processing unit performance. Another series of benchmark tests was performed to determine the effectiveness of various filtering techniques in reducing the negative impact of noisy input terms. Preliminary user tests confirmed our expectation that the Hopfield net algorithm is potentially useful as an associative memory technique to improve document recall and precision by solving discrepancies between indexer vocabularies and end-user vocabularies.",
author = "Hsinchun Chen and Yin Zhang and Houston, {Andrea L.}",
year = "1998",
language = "English (US)",
volume = "24",
pages = "3--18",
journal = "Journal of Information Science",
issn = "0165-5515",
publisher = "SAGE Publications Ltd",
number = "1",

}

TY - JOUR

T1 - Semantic indexing and searching using a Hopfield net

AU - Chen, Hsinchun

AU - Zhang, Yin

AU - Houston, Andrea L.

PY - 1998

Y1 - 1998

N2 - This paper presents a neural network approach to document semantic indexing. A Hopfield net algorithm was used to simulate human associative memory for concept exploration in the domain of computer science and engineering. INSFEC, a collection of more than 320,000 document abstracts from leading journals, was used as the document testbed. Benchmark tests confirmed that three parameters (maximum number of activated nodes, ε - maximum allowable error, and maximum number of iterations] were useful in positively influencing network convergence behavior without negatively impacting central processing unit performance. Another series of benchmark tests was performed to determine the effectiveness of various filtering techniques in reducing the negative impact of noisy input terms. Preliminary user tests confirmed our expectation that the Hopfield net algorithm is potentially useful as an associative memory technique to improve document recall and precision by solving discrepancies between indexer vocabularies and end-user vocabularies.

AB - This paper presents a neural network approach to document semantic indexing. A Hopfield net algorithm was used to simulate human associative memory for concept exploration in the domain of computer science and engineering. INSFEC, a collection of more than 320,000 document abstracts from leading journals, was used as the document testbed. Benchmark tests confirmed that three parameters (maximum number of activated nodes, ε - maximum allowable error, and maximum number of iterations] were useful in positively influencing network convergence behavior without negatively impacting central processing unit performance. Another series of benchmark tests was performed to determine the effectiveness of various filtering techniques in reducing the negative impact of noisy input terms. Preliminary user tests confirmed our expectation that the Hopfield net algorithm is potentially useful as an associative memory technique to improve document recall and precision by solving discrepancies between indexer vocabularies and end-user vocabularies.

UR - http://www.scopus.com/inward/record.url?scp=0031698261&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031698261&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0031698261

VL - 24

SP - 3

EP - 18

JO - Journal of Information Science

JF - Journal of Information Science

SN - 0165-5515

IS - 1

ER -