• Categories

  • Most Popular Questions

  • Recently Viewed Questions

  • Recent Answers

    How To Make a Digita… on What does a frequency of 100 H…
    Daigrepont on Can an earthquake cause air tu…
    Benedict on How did God come into exi…
    joshua on How does the human body g…
    Ian on How did God come into exi…
  • Recent Questions

  • Blog Stats

    • 2,212,389 hits
  • Visitors since 11-3-08

    counter create hit
  • Terms and Conditions

  • Warning

    We are doing maintenance on this site, so some posts may disappear for a short time. Sorry. Normal service will soon be resumed...
  • Pages

  • March 2008
    M T W T F S S
    « Feb   Apr »
     12
    3456789
    10111213141516
    17181920212223
    24252627282930
    31  
  • Archives

  • Meta

How do internet search engines work?

How do internet search engines find the most relevant topic to your search?
Dan Richards from London (Age 25-34)

Advertisements

4 Responses

  1. Different search engines have their own individual ways of working out what is “relevant, and most of them keep the exact techniques used secret to prevent people from fiddling the system to make a particular page appear more relevant than it is. A simple way of finding relevance is just to look for pages that contain the keywords you are searching for and words that are known to be related to them (using something like a thesaurus). This is too easy to fake however and search engines now tend to use mechanisms which take into account the number of links leading to a page, so that a page with the right keywords and lots of links leading to it would be regarded as more relevant than one with fewer links pointing at it. The big engines such as Google are perpetually tuning their algorithms so as to keep ahead of tricks people develop to boost their position in searches.

  2. Take a look at this series of articles by Tim Bray: http://www.tbray.org/ongoing/When/200x/2003/07/30/OnSearchTOC

    He walks you through most of the technology involved in building a fairly simple search engine.

  3. Tim Bray’s series “On Search” on his blog is the best public resource I know of that answers this question, from an industry insider:

    http://www.tbray.org/ongoing/When/200x/2003/07/30/OnSearchTOC

    Steve Crossan works for Google.

  4. As Lindsay says, there are lots of ways to work out “relevance” and the details are often quite difficult to come by. There’s a good paper on the Google approach here:

    http://infolab.stanford.edu/~backrub/google.html

    The essential idea behind this is that a website in the search engine is higher in the rankings if other people on the internet find it useful by “linking to it” from their website.

    I’d suggest that this measure of relevancy seems to work well, but I can see other ways that search engines might calculate relevancy in the future which may be even better. For example, something may be more relevant if my friends like it, or if I am closer to it. I would like to search for “Pizza” and find a restaurant which is close that my friends like. I think as engineers and scientists we’re just starting to explore what relevance really means, and how we can deliver it.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: