permalink

1

Google Future Search Techniques

Google LogoI came across one interesting article at seo by the sea where it says Google has submitted 6 new patents for personalized search. You can say it as an intelligent search if you are not worried about your privacy. One part of the patent filings goes into detail on some possibilities. Here are some details of how a search might be modified by this personalization system: 

  • A search query is received by the search engine which runs the query against the document repository.
  • After the results are received, they are adjusted based upon information from the user’s history.
  • Results that are present in a user’s history could be added to the results presented to the user.
  • The order of the search results can be adjusted in accordance with a history score and/or any user-modified result score.
  • Scores like the search result score and the history score can be combined and with search results is reordered displayed based on that combined score.
  • A searcher may be shown an indication showing previously visited pages, including things like the date and time a page was previously visited and the number of times that the user has visited the site within a certain period of time.
  • A certain number of the most highly ranked results that a searcher has previously visited could be displayed in a area above the search results, or in another section of the page, or even in a separate window.
  • Those previously visited pages may be ordered based upon different ranking criteria, such as a history score, pagerank, time of last access, number of accesses, or others.
  • In addition to previously visited pages, a user’s previous query sessions and/or session groups may also be shown along with the search results and ordered as described above.
  • Search results which a searcher has seen in the page, and which have been clicked upon can be boosted higher in the set of search results.
  • A user’s browsing activities can be considered in addition to, or in lieu of, past presentation and click through of particular search results. For example, if a location was previously visited by a user, it could have its score boosted, based upon the number of times the user has visited the location. So, a site that is bookmarked and visited frequently may rank higher than one which has been searched for and selected in search results.
  • Search results that were previously presented to searchers but not clicked through could be demoted in the results.
  • As an alternative, instead of reordering search results, a history score could be used to determining whether a result is shown with some type of visual indicator such as a change of color or through the use of highlighting.
  • The order of some pages showing in search results could be boosted higher if the searcher has visited or clicked on results from related sites or pages.
  • The amount of times that past queries have retrieved a certain set of results, and how long ago those happened, could also affect the placement of a document in search results.
  • A searcher’s history could be used to identify additional search results. For example, results in a searcher’s history not appearing in a set of results but which were retrieved by similar queries could be added to the results. It’s possible that these might be displayed in a different area of the screen than the initially identified search results.
  • Additional queries might be suggested, possibly even similar queries that were previously submitted by the user. The similarity of these additional queries might be calculated a number of ways, such as by edit distance, stemming operations, correction of obviously misspelled words, semantic mapping, similarity of the retrieved document sets, and others.
  • Queries could also be suggested from the user’s history which were submitted immediately following or preceding the query at issue, in the same query sessions or session group from that history.
  • Queries could be run from different document source repositories. For example, if a searcher runs a web search, and they’ve done a similar search for a product review, they may be given the option of running the query in a product review repository.

From the above points you can say every click and every mouse move on Google website is logged. In the future they will do like we have personalized the website template based on your search :-D. I think with the help of orkut.com they will get to know what kind of friends I have or try to get, my environment, my style, etc etc. Now they have bought youtube so more tracking for Google ;-). After watching the video Cutting Edge 04-07-06 Google Behind The Screen I have stopped using my gmail account and I am slowly moving away from it, as I am more worried about my privacy and me being monitored all the time :-(.

1 Comment