The BINGO! focused crawler: from bookmarks to archetypes

Author

Sizov, Sergej ; Siersdorfer, Stefan ; Theobald, Martin ; Weikum, Gerhard

Author_Institution

Saarlandes Univ., Saarbrucken, Germany

fYear

2002

fDate

2002

Firstpage

337

Lastpage

338

Abstract

The BINGO! system implements an approach to focused crawling that aims to overcome the limitations of the initial training data. To this end, BINGO! identifies, among the crawled and positively classified documents of a topic, characteristic "archetypes" and uses them for periodically re-training the classifier; this way the crawler is dynamically adapted based on the most significant documents seen so far. Two kinds of archetypes are considered: good authorities as determined by employing Kleinberg\´s link analysis algorithm, and documents that have been automatically classified with high confidence using a linear SVM classifier

Keywords

classification; hypermedia markup languages; BINGO! focused crawler; Kleinberg link analysis algorithm; archetypes; best URLs; bookmarks; crawl frontier; linear SVM classifier; positively classified documents; re-training; Costs; Crawlers; Humans; Ontologies; Search engines; Support vector machine classification; Support vector machines; Training data; Uniform resource locators; World Wide Web;

fLanguage

English

Publisher

ieee

Conference_Titel

Data Engineering, 2002. Proceedings. 18th International Conference on

Conference_Location

San Jose, CA

ISSN

1063-6382

Print_ISBN

0-7695-1531-2

Type

conf

DOI

10.1109/ICDE.2002.994746

Filename

994746