• DocumentCode
    3103283
  • Title

    Describing Web pages better in search results

  • Author

    Deepak, P. ; John, Jyothi

  • Author_Institution
    Model Eng. Coll., Kochi, India
  • fYear
    2004
  • fDate
    19-23 April 2004
  • Firstpage
    629
  • Lastpage
    630
  • Abstract
    This work presents an algorithm for generating internet Web page descriptions automatically in situations where the need is for a lot of information within a few sentences, a typical example being the descriptions of Web pages in search results. Such descriptions often prove to be very useful aids for navigation. Many search engines provide extracts from the page as a description. But studies say that extracts from pages do not usually serve the cause. A page is often described by other pages better that by itself. This work presents an algorithm, which retrieves descriptions about a page from those linking to it and ranks them. The descriptions generated on test sets have been compared to those generated by acclaimed search engines. It may well be noted that Google (http://www.google.com) and many other search engines have not publicized the algorithm that they use.
  • Keywords
    Internet; information retrieval; online front-ends; search engines; Google Web site; Web page description; information navigation; search engine; Data mining; Educational institutions; Humans; Joining processes; Roads; Search engines; Testing; Uniform resource locators; Upper bound; Web pages;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Information and Communication Technologies: From Theory to Applications, 2004. Proceedings. 2004 International Conference on
  • Print_ISBN
    0-7803-8482-2
  • Type

    conf

  • DOI
    10.1109/ICTTA.2004.1307922
  • Filename
    1307922