Visualizing the Semantic Web

pp 151-167

Recommender Systems for the Web

  • Joseph A. Konstan
  • , John T. Riedl

* Final gross prices may vary according to local VAT.

Get Access


As the web rapidly evolved into an immense repository of content, human users discovered that they could no longer effectively identify the content of most interest to them. Several approaches developed for improving our ability to find content. Syntactic search engines helped index and rapidly scan millions of pages for keywords, but we quickly learned that the amount of content with matching keywords was still too high. Semantic annotation helps assist automated (or computer-assisted) processing of content to better identify the real contents of pages. A Semantic web would help people differentiate between articles on “china” plates and articles about “China” the country. Recommender systems tried a different approach — relying on the collected efforts of a community of users to assess the quality and importance of contents. For example, a web page recommender may identify that a particular page is popular overall, or better yet, popular among people with tastes like yours.