Mining Highly Authoritative Web Resources for One-Stop Learning

• Published in: IEEE/WIC/ACM International Conference on web intelligence pp. 289 - 292
• Main Authors: Lim, SeungJin, Ko, Youngrae
• Format: Conference Proceeding
• Language: English
• Published: Washington, DC, USA IEEE Computer Society 19.09.2005; IEEE
• Series: ACM Conferences
• Subjects: Computer science; Delay; HTML; Information systems > Information retrieval; Information systems > Information retrieval > Retrieval models and ranking; Information systems > Information retrieval > Retrieval tasks and goals; Information systems > Information storage systems; Information systems > Information systems applications > Data mining; Learning systems; Robustness; Search engines; Web pages
• ISBN: 076952415X; 9780769524153
• DOI: 10.1109/WI.2005.97

The convenience of the Web equipped with automatic search engines attracts "focused learners" for learning about a new subject of interest. The resources recommended by a search engine are, however, often a collection of links to other resources, or commercial-driven, irrelevant, misleading pages. Subsequently, the learner needs to manually click through numerous pages to find quality resources. This paper proposes an approach to a new problem of mining the most suitable resources for one-stop learning, called "highly authoritative resources." The experimental results using top search results from Google and Yahoo for various subjects show that the proposed algorithm is highly effective both in quality and time.