Fast adapting value estimation-based hybrid architecture for searching the world-wide web

István Kókai, A. Lőrincz

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

The slogan that information is power has undergone a slight change. Today, information updating is in the focus of interest. The largest source of information is the world-wide web. Fast search methods are in need for this enormous source. In this paper a hybrid architecture that combines soft support vector classification and reinforcement learning for value estimation is introduced for the evaluation of a link (a document) and its neighboring links (or documents), called the context of a document. The method is motivated by (i) large differences between such contexts on the web, (ii) the facilitation of goal oriented search using context classifiers, and (iii) attractive fast adaptation properties, that could counteract diversity of web environments. We demonstrate that value estimation-based fast adaptation offers considerable improvement over other known search methods.

Original languageEnglish
Pages (from-to)11-23
Number of pages13
JournalApplied Soft Computing Journal
Volume2
Issue number1
DOIs
Publication statusPublished - 2002

Fingerprint

World Wide Web
Reinforcement learning
Classifiers

Keywords

  • Fast adaptation
  • Internet
  • Reinforcement learning
  • Search
  • Small world
  • SVM

ASJC Scopus subject areas

  • Software
  • Computer Science Applications
  • Artificial Intelligence

Cite this

Fast adapting value estimation-based hybrid architecture for searching the world-wide web. / Kókai, István; Lőrincz, A.

In: Applied Soft Computing Journal, Vol. 2, No. 1, 2002, p. 11-23.

Research output: Contribution to journalArticle

@article{433059070f4842279c1168a0a6880ecf,
title = "Fast adapting value estimation-based hybrid architecture for searching the world-wide web",
abstract = "The slogan that information is power has undergone a slight change. Today, information updating is in the focus of interest. The largest source of information is the world-wide web. Fast search methods are in need for this enormous source. In this paper a hybrid architecture that combines soft support vector classification and reinforcement learning for value estimation is introduced for the evaluation of a link (a document) and its neighboring links (or documents), called the context of a document. The method is motivated by (i) large differences between such contexts on the web, (ii) the facilitation of goal oriented search using context classifiers, and (iii) attractive fast adaptation properties, that could counteract diversity of web environments. We demonstrate that value estimation-based fast adaptation offers considerable improvement over other known search methods.",
keywords = "Fast adaptation, Internet, Reinforcement learning, Search, Small world, SVM",
author = "Istv{\'a}n K{\'o}kai and A. Lőrincz",
year = "2002",
doi = "10.1016/S1568-4946(02)00025-X",
language = "English",
volume = "2",
pages = "11--23",
journal = "Applied Soft Computing",
issn = "1568-4946",
publisher = "Elsevier BV",
number = "1",

}

TY - JOUR

T1 - Fast adapting value estimation-based hybrid architecture for searching the world-wide web

AU - Kókai, István

AU - Lőrincz, A.

PY - 2002

Y1 - 2002

N2 - The slogan that information is power has undergone a slight change. Today, information updating is in the focus of interest. The largest source of information is the world-wide web. Fast search methods are in need for this enormous source. In this paper a hybrid architecture that combines soft support vector classification and reinforcement learning for value estimation is introduced for the evaluation of a link (a document) and its neighboring links (or documents), called the context of a document. The method is motivated by (i) large differences between such contexts on the web, (ii) the facilitation of goal oriented search using context classifiers, and (iii) attractive fast adaptation properties, that could counteract diversity of web environments. We demonstrate that value estimation-based fast adaptation offers considerable improvement over other known search methods.

AB - The slogan that information is power has undergone a slight change. Today, information updating is in the focus of interest. The largest source of information is the world-wide web. Fast search methods are in need for this enormous source. In this paper a hybrid architecture that combines soft support vector classification and reinforcement learning for value estimation is introduced for the evaluation of a link (a document) and its neighboring links (or documents), called the context of a document. The method is motivated by (i) large differences between such contexts on the web, (ii) the facilitation of goal oriented search using context classifiers, and (iii) attractive fast adaptation properties, that could counteract diversity of web environments. We demonstrate that value estimation-based fast adaptation offers considerable improvement over other known search methods.

KW - Fast adaptation

KW - Internet

KW - Reinforcement learning

KW - Search

KW - Small world

KW - SVM

UR - http://www.scopus.com/inward/record.url?scp=6344271846&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=6344271846&partnerID=8YFLogxK

U2 - 10.1016/S1568-4946(02)00025-X

DO - 10.1016/S1568-4946(02)00025-X

M3 - Article

AN - SCOPUS:6344271846

VL - 2

SP - 11

EP - 23

JO - Applied Soft Computing

JF - Applied Soft Computing

SN - 1568-4946

IS - 1

ER -