Keyword spices: A new method for building domain-specific web search engines

Satoshi Oyama, Takashi Kokubo, Toru Ishida, Teruhiro Yamada, Yasuhiko Kitamura

Research output: Contribution to journalConference article

20 Citations (Scopus)

Abstract

This paper presents a new method for building domain-specific web search engines. Previous methods eliminate irrelevant documents from the pages accessed using heuristics based on human knowledge about the domain in question. Accordingly, they are hard to build and can not be applied to other domains. The keyword spice method, in contrast, improves search performance by adding domain-specific keywords, called keyword spices, to the user's input query; the modified query is then forwarded to a general-purpose search engine. Keyword spices can be effectively discovered automatically from web documents allowing us to build high quality domain-specific search engines in various domains without requiring the collection of heuristic knowledge. We describe a machine learning algorithm, which is a type of decision-tree learning algorithm, that can extract keyword spices. To demonstrate the value of the proposed approach, we conduct experiments in the domain of cooking. The results confirm the excellent performance of our method in terms of both precision and recall.

Original languageEnglish
Pages (from-to)1457-1463
Number of pages7
JournalIJCAI International Joint Conference on Artificial Intelligence
Publication statusPublished - 2001 Dec 1
Externally publishedYes
Event17th International Joint Conference on Artificial Intelligence, IJCAI 2001 - Seattle, WA, United States
Duration: 2001 Aug 42001 Aug 10

Fingerprint

Search engines
Learning algorithms
Cooking
Decision trees
Learning systems
Experiments

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Keyword spices : A new method for building domain-specific web search engines. / Oyama, Satoshi; Kokubo, Takashi; Ishida, Toru; Yamada, Teruhiro; Kitamura, Yasuhiko.

In: IJCAI International Joint Conference on Artificial Intelligence, 01.12.2001, p. 1457-1463.

Research output: Contribution to journalConference article

Oyama, Satoshi ; Kokubo, Takashi ; Ishida, Toru ; Yamada, Teruhiro ; Kitamura, Yasuhiko. / Keyword spices : A new method for building domain-specific web search engines. In: IJCAI International Joint Conference on Artificial Intelligence. 2001 ; pp. 1457-1463.
@article{ec2b1eab8c5f47e0a1fb17d4f0e71521,
title = "Keyword spices: A new method for building domain-specific web search engines",
abstract = "This paper presents a new method for building domain-specific web search engines. Previous methods eliminate irrelevant documents from the pages accessed using heuristics based on human knowledge about the domain in question. Accordingly, they are hard to build and can not be applied to other domains. The keyword spice method, in contrast, improves search performance by adding domain-specific keywords, called keyword spices, to the user's input query; the modified query is then forwarded to a general-purpose search engine. Keyword spices can be effectively discovered automatically from web documents allowing us to build high quality domain-specific search engines in various domains without requiring the collection of heuristic knowledge. We describe a machine learning algorithm, which is a type of decision-tree learning algorithm, that can extract keyword spices. To demonstrate the value of the proposed approach, we conduct experiments in the domain of cooking. The results confirm the excellent performance of our method in terms of both precision and recall.",
author = "Satoshi Oyama and Takashi Kokubo and Toru Ishida and Teruhiro Yamada and Yasuhiko Kitamura",
year = "2001",
month = "12",
day = "1",
language = "English",
pages = "1457--1463",
journal = "IJCAI International Joint Conference on Artificial Intelligence",
issn = "1045-0823",

}

TY - JOUR

T1 - Keyword spices

T2 - A new method for building domain-specific web search engines

AU - Oyama, Satoshi

AU - Kokubo, Takashi

AU - Ishida, Toru

AU - Yamada, Teruhiro

AU - Kitamura, Yasuhiko

PY - 2001/12/1

Y1 - 2001/12/1

N2 - This paper presents a new method for building domain-specific web search engines. Previous methods eliminate irrelevant documents from the pages accessed using heuristics based on human knowledge about the domain in question. Accordingly, they are hard to build and can not be applied to other domains. The keyword spice method, in contrast, improves search performance by adding domain-specific keywords, called keyword spices, to the user's input query; the modified query is then forwarded to a general-purpose search engine. Keyword spices can be effectively discovered automatically from web documents allowing us to build high quality domain-specific search engines in various domains without requiring the collection of heuristic knowledge. We describe a machine learning algorithm, which is a type of decision-tree learning algorithm, that can extract keyword spices. To demonstrate the value of the proposed approach, we conduct experiments in the domain of cooking. The results confirm the excellent performance of our method in terms of both precision and recall.

AB - This paper presents a new method for building domain-specific web search engines. Previous methods eliminate irrelevant documents from the pages accessed using heuristics based on human knowledge about the domain in question. Accordingly, they are hard to build and can not be applied to other domains. The keyword spice method, in contrast, improves search performance by adding domain-specific keywords, called keyword spices, to the user's input query; the modified query is then forwarded to a general-purpose search engine. Keyword spices can be effectively discovered automatically from web documents allowing us to build high quality domain-specific search engines in various domains without requiring the collection of heuristic knowledge. We describe a machine learning algorithm, which is a type of decision-tree learning algorithm, that can extract keyword spices. To demonstrate the value of the proposed approach, we conduct experiments in the domain of cooking. The results confirm the excellent performance of our method in terms of both precision and recall.

UR - http://www.scopus.com/inward/record.url?scp=84880913144&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84880913144&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:84880913144

SP - 1457

EP - 1463

JO - IJCAI International Joint Conference on Artificial Intelligence

JF - IJCAI International Joint Conference on Artificial Intelligence

SN - 1045-0823

ER -