System and method for multi-dimensional personization of search results

Abstract

In various example embodiments, a system and method for personalization of search results are provided. In example embodiments, a query that triggers a search of a data storage device of a publication system that comprises a plurality of publications is received and a search performed to determine a result set of publications. Optimization preferences of the user are accessed and applied to the result set obtained based on the query to generate a personalized result set. The personalized result set is presented on a user interface of a device of the user.

1 . (canceled) 
     
     
         2 . A method comprising:
 receiving a query from a user that triggers a search of a data storage device of a publication system that comprises a plurality of publications;   performing the search of the data storage device to identify a result set of publications that match the query;   accessing optimization preferences of the user, the optimization preferences including a plurality of interrelated optimization factors and a value assigned to each of the interrelated optimization factors;   determining a factor score corresponding to each of the interrelated optimization factors for each publication in the result set;   applying, using at least one processor of a machine, the optimization preferences to the factor score corresponding to each of the interrelated optimization factors for each publication in the result set to determine a unifying score for each publication in the result set;   deriving a personalized result set of publications based on a selection of publications in the result set having highest unifying scores; and   causing presentation of the personalized result set on a user interface of a device of the user.   
     
     
         3 . The method of  claim 2 , wherein the value assigned to each of the interrelated optimization factors corresponds to a coefficient that is applied to a corresponding factor score. 
     
     
         4 . The method of  claim 3 , wherein the unifying score for each publication comprises a summation of each coefficient times the corresponding factor score for each publication. 
     
     
         5 . The method of  claim 2 , further comprising adding at least one perturbation match to the personalized result set, the at least one perturbation match being a match that is lacking in one or more optimization factors. 
     
     
         6 . The method of  claim 2 , wherein the determining the factor score comprises determining a diversity score for each publication of the result set, the determining of the diversity score comprising:
 selecting a most popular match and placing the most popular match in a potential set;   continually determining a next most divergent match from results in the potential set and placing the next most divergent match in the potential set until a limit is reached;   assigning a highest diversity score to the most popular match; and   assigning proportionally decreasing diversity scores for a remainder of the matches in the potential set.   
     
     
         7 . The method of  claim 2 , further comprising:
 causing presentation of a preference input interface that allows the user to establish the optimization preferences, the preference input interface comprising a plurality of sliding scales corresponding to each of the plurality of interrelated optimization factors, the value assigned to each of the interrelated optimization factors being inputted using the sliding scale; and   receiving the optimization preferences via the preference input interface.   
     
     
         8 . The method of  claim 2 , further comprising:
 Causing presentation of a preference input interface that allows the user to establish the optimization preferences, the preference input interface comprising a radar chart, each node of the radar chart corresponding to an interrelated optimization factor of the plurality of interrelated optimization factors, the value assigned to each of the plurality of interrelated optimization factors being inputted by adjusting the nodes of the radar chart; and   receiving the optimization preferences via the preference input interface.   
     
     
         9 . The method of  claim 2 , further comprising:
 examining past searches and transactions associated with an account of the user to determine patterns indicative of user preferences;   deriving suggested optimization preferences based on the user preferences; and   presenting the suggested optimization preference as the optimization preferences, the user having an ability to adjust the suggested optimization preferences.   
     
     
         10 . A system comprising:
 one or more hardware processors configured to perform operations comprising:
 receiving a query from a user that triggers a search of a data storage device of a publication system that comprises a plurality of publications; 
 performing the search of the data storage device to identify a result set of publications that match the query; 
 accessing optimization preferences of the user, the optimization preferences including a plurality of interrelated optimization factors and a value assigned to each of the interrelated optimization factors; 
 determining a factor score corresponding to each of the interrelated optimization factors for each publication in the result set; 
 applying the optimization preferences to the factor score corresponding to each of the interrelated optimization factors for each publication in the result set to determine a unifying score for each publication in the result set; 
 deriving a personalized result set of publications based on a selection of publications in the result set having highest unifying scores; and 
 causing presentation of the personalized result set on a user interface of a device of the user. 
   
     
     
         11 . The system of  claim 10 , wherein the operations further comprise adding at least one perturbation match to the personalized result set, the at least one perturbation match being a match that is lacking in one or more optimization factors. 
     
     
         12 . The system of  claim 10 , wherein the determining the factor score comprises determining a diversity score for each publication of the result set, the determining of the diversity score comprising:
 selecting a most popular match and placing the most popular match in a potential set;   continually determining a next most divergent match from results in the potential set and placing the next most divergent match in the potential set until a limit is reached;   assigning a highest diversity score to the most popular match; and   assigning proportionally decreasing diversity scores for a remainder of the matches in the potential set.   
     
     
         13 . The system of  claim 10 , wherein the operations further comprise:
 causing presentation of a preference input interface that allows the user to establish the optimization preferences, the preference input interface comprising a plurality of sliding scales corresponding to each of the plurality of interrelated optimization factors, the value assigned to each of the interrelated optimization factors being inputted using the sliding scale; and   receiving the optimization preferences via the preference input interface.   
     
     
         14 . The system of  claim 10 , wherein the operations further comprise:
 causing presentation of a preference input interface that allows the user to establish the optimization preferences, the preference input interface comprising a radar chart, each node of the radar chart corresponding to an interrelated optimization factor of the plurality of interrelated optimization factors, the value assigned to each of the plurality of interrelated optimization factors being inputted by adjusting the nodes of the radar chart; and   receiving the optimization preferences via the preference input interface.   
     
     
         15 . The system of  claim 10 , wherein the operations further comprise:
 examining past searches and transactions associated with an account of the user to determine patterns indicative of user preferences;   deriving suggested optimization preferences based on the user preferences; and   presenting the suggested optimization preference as the optimization preferences, the user having an ability to adjust the suggested optimization preferences.   
     
     
         16 . A machine-readable medium having no transitory signals and storing instructions which, when executed by the at least one processor of a machine, causes the machine to perform operations comprising:
 receiving a query from a user that triggers a search of a data storage device of a publication system that comprises a plurality of publications;   performing the search of the data storage device to identify a result set of publications that match the query;   accessing optimization preferences of the user, the optimization preferences including a plurality of interrelated optimization factors and a value assigned to each of the interrelated optimization factors;   determining a factor score corresponding to each of the interrelated optimization factors for each publication in the result set;   applying the optimization preferences to the factor score corresponding to each of the interrelated optimization factors for each publication in the result set to determine a unifying score for each publication in the result set;   deriving a personalized result set of publications based on a selection of publications in the result set having highest unifying scores; and   causing presentation of the personalized result set on a user interface of a device of the user.   
     
     
         17 . The machine-readable medium of  claim 16 , wherein the operations further comprise adding at least one perturbation match to the personalized result set, the at least one perturbation match being a match that is lacking in one or more optimization factors. 
     
     
         18 . The machine-readable medium of  claim 16 , wherein the determining the factor score comprises determining a diversity score for each publication of the result set, the determining of the diversity score comprising:
 selecting a most popular match and placing the most popular match in a potential set;   continually determining a next most divergent match from results in the potential set and placing the next most divergent match in the potential set until a limit is reached;   assigning a highest diversity score to the most popular match; and   assigning proportionally decreasing diversity scores for a remainder of the matches in the potential set.   
     
     
         19 . The machine-readable medium of  claim 16 , wherein the operations further comprise:
 causing presentation of a preference input interface that allows the user to establish the optimization preferences, the preference input interface comprising a plurality of sliding scales corresponding to each of the plurality of interrelated optimization factors, the value assigned to each of the interrelated optimization factors being inputted using the sliding scale; and   receiving the optimization preferences via the preference input interface.   
     
     
         20 . The machine-readable medium of  claim 16 , wherein the operations further comprise:
 causing presentation of a preference input interface that allows the user to establish the optimization preferences, the preference input interface comprising a radar chart, each node of the radar chart corresponding to an interrelated optimization factor of the plurality of interrelated optimization factors, the value assigned to each of the plurality of interrelated optimization factors being inputted by adjusting the nodes of the radar chart; and   receiving the optimization preferences via the preference input interface.   
     
     
         21 . The machine-readable medium of  claim 16 , wherein the operations further comprise:
 examining past searches and transactions associated with an account of the user to determine patterns indicative of user preferences;   deriving suggested optimization preferences based on the user preferences; and   presenting the suggested optimization preference as the optimization preferences, the user having an ability to adjust the suggested optimization preferences.