Classifying information sender of web documents

Yoshikiyo Kato, Sadao Kurohashi, Kentaro Inui

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

As more diverse information becomes available on the Web in terms of variety and quality, the need for support beyond current search engine technology for judging credibility of information grows. In this paper, we report a method for classifying information sender of web documents as a first step toward assessing information credibility based on characteristics of the information sender. We propose a method which identify the sender class of a given web document with relatively few information, i.e. its URL, title, and the title of its top page. The evaluation shows that with the proposed method we can classify the sender of web documents at above 70% accuracy with 30% coverage for unseen data.

Original languageEnglish
Title of host publicationProceedings of the First Workshop on Information Credibility on the Web (WICOW) in Conjunction with the 21st Annual Conference of the Japanese Society for Artificial Intelligence, 2007
PublisherJapanese Society for Artificial Intelligence
ISBN (Print)4915905284, 9784915905285
Publication statusPublished - 2007 Jan 1
Externally publishedYes
Event1st Workshop on Information Credibility on the Web, WICOW 2007 - Miyazaki, Japan
Duration: 2007 Jun 192007 Jun 19

Publication series

NameProceedings of the First Workshop on Information Credibility on the Web (WICOW) in Conjunction with the 21st Annual Conference of the Japanese Society for Artificial Intelligence, 2007

Other

Other1st Workshop on Information Credibility on the Web, WICOW 2007
CountryJapan
CityMiyazaki
Period07/6/1907/6/19

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications

Fingerprint Dive into the research topics of 'Classifying information sender of web documents'. Together they form a unique fingerprint.

Cite this