How Does ART Ensure the First Results Are Not Falsely Promoted by Users Clicking Them Solely Because They Are the First Results?

ART evaluates user visits as a whole and allocates additional weight (i.e., gives a higher score) to the last clicked item rather than the previously opened items.

A user performs a query and inspects the returned results:

  • The user clicks the top 2 items, but these do not answer their inquiry.

  • The user clicks the three following results, and then closes the search page.

  • ART assumes the fifth result is the best result (i.e., answer the user inquiry) since the click on this result is the last user action associated with the search event.

  • As the ART model is rebuilt over time, the fifth result climbs the result list based on the additional weight acquired from being the last opened item.

  • ART boosts 5 results to make sure that results 5 to 10 are based on standard index-ranking, thus allowing those results to move up in the result list.

  • To boost results, ART does not simply use the result popularity, but takes several other factors into account to also prevent positional bias (also know as high ranking result bias).