Combining segments of users into vertically indexed super-segments

Abstract

Techniques are described herein for combining segments of users into super-segments. The interests of the segments in online ad campaigns in which they participate are compared to determine subsets of the segments whose interests are correlated to an extent that exceeds a designated threshold. The segments in each subset are combined to provide a respective super-segment. Each online ad campaign corresponds to a specified category, such as telecommunications, auto, finance, retail, entertainment, travel, etc. For each super-segment, the cumulative interest of the segments therein for each category are determined The cumulative interests for each super-segment are used to determine which super-segment(s) are to be targeted for receiving a specified online advertisement. For instance, the cumulative interests of the respective super-segments in the category most closely related to a subject matter with which the specified online advertisement is associated may be compared to determine which super-segment(s) are to be targeted.

1 - 20 . (canceled) 
     
     
         21 . A system, comprising:
 a processor; and   a memory device configured to store instructions executable by the processor to:   determine interest values of respective segments of users that are targeted by an online ad campaign of a plurality of online ad campaigns;   remove the online ad campaign from the plurality of online ad campaigns, where a value associated with a summation of the interest values is less than or equal to a designated threshold;   remove segments of the respective segments, which correspond to the removed online ad campaign; and   target remaining segments of the users according to remaining online ad campaigns of the plurality of online ad campaigns.   
     
     
         22 . The system of claim  1 , wherein the instructions are further executable by the processor to determine super-segments based on the interest values, a super-segment including a subset of the respective segments. 
     
     
         23 . The system of claim  2 , wherein the super segments are un-correlated. 
     
     
         24 . The system of claim  2 , wherein the subset of the respective segments includes interest values correlated with the remaining online ad campaigns. 
     
     
         25 . The system of claim  4 , wherein the correlation is to an extent that exceeds a designated threshold. 
     
     
         26 . The system of claim  4 , wherein the interest values correlated with the remaining online ad campaigns indicate an interest in a category specific to the subset of the respective segments. 
     
     
         27 . The system of claim  6 , wherein the instructions are further executable by the processor to determine which of the super-segments are to be targeted by an online advertisement according to the category specific to the subset of the respective segments. 
     
     
         28 . A method, comprising:
 for each online ad campaign of a plurality of online ad campaigns, determining, by a processor, interest values of respective segments of users that are targeted by the online ad campaign;   removing the online ad campaign from the plurality of online ad campaigns, where a value associated with a summation of the interest values is less than a designated threshold;   removing segments of the respective segments, which correspond to the removed online ad campaign; and   targeting remaining segments of the users according to remaining online ad campaigns of the plurality of online ad campaigns.   
     
     
         29 . The method of claim  8 , further comprising: determining super-segments based on the interest values, a super-segment including a subset of the respective segments. 
     
     
         30 . The method of claim  9 , wherein the super segments are un-correlated. 
     
     
         31 . The method of claim  9 , wherein the subset of the respective segments includes interest values correlated with the remaining online ad campaigns. 
     
     
         32 . The method of claim  11 , wherein the correlation is to an extent that exceeds a designated threshold. 
     
     
         33 . The method of claim  11 , wherein the interest values correlated with the remaining online ad campaigns indicate an interest in a category specific to the subset of the respective segments. 
     
     
         34 . The method of claim  13 , further comprising: determining which of the super-segments are to be targeted by an online advertisement according to the category specific to the subset of the respective segments. 
     
     
         35 . A non-transitory computer readable medium, comprising:
 instructions executable to determine interest values of respective segments of users that are targeted by an online ad campaign of a plurality of online ad campaigns;   instructions executable to remove the online ad campaign from the plurality of online ad campaigns, where a value associated with a summation of the interest values is less than or equal to a designated threshold;   instructions executable to remove segments of the respective segments, which correspond to the removed online ad campaign; and   instructions executable to output to a display device remaining online ad campaigns of the plurality of online ad campaigns.   
     
     
         36 . The medium of claim  15 , further comprising instructions executable to determine super-segments based on the interest values, a super-segment including a subset of the respective segments. 
     
     
         37 . The medium of claim  16 , wherein the super segments are un-correlated. 
     
     
         38 . The medium of claim  16 , wherein the subset of the respective segments includes interest values correlated with the remaining online ad campaigns. 
     
     
         39 . The medium of claim  18 , wherein the correlation is to an extent that exceeds a designated threshold. 
     
     
         40 . The medium of claim  18 , further comprising instructions executable to determine which of the super-segments are to be targeted by an online advertisement according to a category specific to the subset of the respective segments.