Topic and time based media affinity estimation

Abstract

An affinity server estimates an affinity between two different time based media events (e.g., TV, radio, social media content stream), between a time based media event and a specific topic, or between two different topics, where the affinity score represents an intersection between the populations of social media users who have authored social media content items regarding the two different events and/or topics. The affinity score represents an estimation of the real world affinity between the real world population of people who have an interest in both time based media events, both topics, or in a time based media event and a topic. One possible threshold for including a social media user in a population may be based on a confidence score that indicates the confidence that one or more social media content items authored by the social media user are relevant to the topic or event in question.

1 . (canceled) 
     
     
         2 . A method, comprising:
 accessing an event repository comprising a time based media event;   aggregating a first population of social media users, the aggregating comprising:
 accessing a content repository comprising social media content items authored by respective social media users, 
 determining a first confidence score indicative of a probability that respective social media content items are relevant to the time based media event, and 
 adding one or more social media users to the first population based on the first confidence scores; and 
   sending content to client devices associated with one or more social media users in the first population.   
     
     
         3 . The method of  claim 2 , further comprising:
 accessing a topic repository comprising a plurality of topics;   aggregating a second population of social media users, the aggregating comprising:
 accessing the content repository comprising social media content items authored by respective social media users, 
 determining a second confidence score indicative of a probability that respective social media content items are relevant to a first topic, and 
 adding one or more social media users to the second population based on the second confidence score; and 
   determining a first affinity score indicative of an affinity by social media users for both the time based media event and the first topic, the affinity score based on an intersection of social media users in both the first population and the second population, and wherein sending the content is based on the first affinity score.   
     
     
         4 . The method of  claim 3 , further comprising:
 aggregating a third population of social media users, the aggregating comprising:
 accessing the content repository comprising social media content items authored by respective social media users, 
 determining a third confidence score indicative of a probability that respective social media content items are relevant to a second topic, and 
 adding one or more social media users to the third population based on the third confidence score; and 
   determining a second affinity score indicative of an affinity by social media users for both the time based media event and the second topic, the second affinity score based on an intersection of social media users in both the first population and the third population,   wherein, sending the content is based on social media users that satisfy a threshold first affinity and threshold second affinity.   
     
     
         5 . The method of  claim 3 , further comprising:
 normalizing the first affinity score to represent a proportion of total social media users relevant to the time based media event;   obtaining an external measure of the number of real world people who have watched the time based media event; and   using a combination of the proportion and the external measure to determine the one or more social media users to send the content.   
     
     
         6 . The method of  claim 3 , wherein determining the first affinity score further comprises:
 weighting a count of intersecting social media users based on how many times each individual social media user in the intersecting population authors a social media content item relevant to the first topic or the time based media event.   
     
     
         7 . The method of  claim 2 , wherein determining that at least one social media content item authored by the social media user is relevant to the first time based media event comprises:
 extracting event features from annotations associated with the time based media event;   extracting social media features from the social media content item; and   determining the confidence score based on a relationship between the event features and social media features.   
     
     
         8 . The method of  claim 2 , wherein aggregating the first population of social media users further comprises:
 filtering social media content users based on one or more filtering criteria, the filtering criteria comprising one or more of: i) social media user demographic information, ii) a content of the social media content items authored by the social media users, or iii) a time of authorship of the social media content items.   
     
     
         9 . A system comprising one or more computers and a memory storing computer program instructions that when executed by the one or more computers causes the one or more computers to perform operations comprising:
 accessing an event repository comprising a time based media event;   aggregating a first population of social media users, the aggregating comprising:
 accessing a content repository comprising social media content items authored by respective social media users, 
 determining a first confidence score indicative of a probability that respective social media content items are relevant to the time based media event, and 
 adding one or more social media users to the first population based on the first confidence scores; and 
   sending content to client devices associated with one or more social media users in the first population.   
     
     
         10 . The system of  claim 9 , wherein the instructions further cause the one or more computers to perform operations comprising:
 accessing a topic repository comprising a plurality of topics;   aggregating a second population of social media users, the aggregating comprising:
 accessing the content repository comprising social media content items authored by respective social media users, 
 determining a second confidence score indicative of a probability that respective social media content items are relevant to a first topic, and 
 adding one or more social media users to the second population based on the second confidence score; and 
   determining a first affinity score indicative of an affinity by social media users for both the time based media event and the first topic, the affinity score based on an intersection of social media users in both the first population and the second population, and wherein sending the content is based on the first affinity score.   
     
     
         11 . The system of  claim 10 , wherein the instructions further cause the one or more computers to perform operations comprising:
 aggregating a third population of social media users, the aggregating comprising:
 accessing the content repository comprising social media content items authored by respective social media users, 
 determining a third confidence score indicative of a probability that respective social media content items are relevant to a second topic, and 
 adding one or more social media users to the third population based on the third confidence score; and 
   determining a second affinity score indicative of an affinity by social media users for both the time based media event and the second topic, the second affinity score based on an intersection of social media users in both the first population and the third population,   wherein, sending the content is based on social media users that satisfy a threshold first affinity and threshold second affinity.   
     
     
         12 . The system of  claim 10 , wherein the instructions further cause the one or more computers to perform operations comprising:
 normalizing the first affinity score to represent a proportion of total social media users relevant to the time based media event;   obtaining an external measure of the number of real world people who have watched the time based media event; and   using a combination of the proportion and the external measure to determine the one or more social media users to send the content.   
     
     
         13 . The system of  claim 10 , wherein determining the first affinity score further comprises:
 weighting a count of intersecting social media users based on how many times each individual social media user in the intersecting population authors a social media content item relevant to the first topic or the time based media event.   
     
     
         14 . The system of  claim 9 , wherein determining that at least one social media content item authored by the social media user is relevant to the first time based media event comprises:
 extracting event features from annotations associated with the time based media event;   extracting social media features from the social media content item; and   determining the confidence score based on a relationship between the event features and social media features.   
     
     
         15 . The system of  claim 9 , wherein aggregating the first population of social media users further comprises:
 filtering social media content users based on one or more filtering criteria, the filtering criteria comprising one or more of: i) social media user demographic information, ii) a content of the social media content items authored by the social media users, or iii) a time of authorship of the social media content items.   
     
     
         16 . One or more non-transitory computer-readable storage media comprising instructions that when executed cause a system of one or more computers to perform operations comprising:
 accessing an event repository comprising a time based media event;   aggregating a first population of social media users, the aggregating comprising:
 accessing a content repository comprising social media content items authored by respective social media users, 
 determining a first confidence score indicative of a probability that respective social media content items are relevant to the time based media event, and 
 adding one or more social media users to the first population based on the first confidence scores; and 
   sending content to client devices associated with one or more social media users in the first population.   
     
     
         17 . The non-transitory computer-readable storage media of  claim 16 , wherein the instructions further cause the system to perform operations comprising:
 accessing a topic repository comprising a plurality of topics;   aggregating a second population of social media users, the aggregating comprising:
 accessing the content repository comprising social media content items authored by respective social media users, 
 determining a second confidence score indicative of a probability that respective social media content items are relevant to a first topic, and 
 adding one or more social media users to the second population based on the second confidence score; and 
   determining a first affinity score indicative of an affinity by social media users for both the time based media event and the first topic, the affinity score based on an intersection of social media users in both the first population and the second population, and wherein sending the content is based on the first affinity score.   
     
     
         18 . The non-transitory computer-readable storage media of  claim 17 , wherein the instructions further cause the system to perform operations comprising:
 aggregating a third population of social media users, the aggregating comprising:
 accessing the content repository comprising social media content items authored by respective social media users, 
 determining a third confidence score indicative of a probability that respective social media content items are relevant to a second topic, and 
 adding one or more social media users to the third population based on the third confidence score; and 
   determining a second affinity score indicative of an affinity by social media users for both the time based media event and the second topic, the second affinity score based on an intersection of social media users in both the first population and the third population,   wherein, sending the content is based on social media users that satisfy a threshold first affinity and threshold second affinity.   
     
     
         19 . The non-transitory computer-readable storage media of  claim 17 , wherein the instructions further cause the system to perform operations comprising:
 normalizing the first affinity score to represent a proportion of total social media users relevant to the time based media event;   obtaining an external measure of the number of real world people who have watched the time based media event; and   using a combination of the proportion and the external measure to determine the one or more social media users to send the content.   
     
     
         20 . The non-transitory computer-readable storage media of  claim 17 , wherein determining the first affinity score further comprises:
 weighting a count of intersecting social media users based on how many times each individual social media user in the intersecting population authors a social media content item relevant to the first topic or the time based media event.   
     
     
         21 . The non-transitory computer-readable storage media of  claim 16 , wherein determining that at least one social media content item authored by the social media user is relevant to the first time based media event comprises:
 extracting event features from annotations associated with the time based media event;   extracting social media features from the social media content item; and   determining the confidence score based on a relationship between the event features and social media features.   
     
     
         22 . The non-transitory computer-readable storage media of  claim 16 , wherein aggregating the first population of social media users further comprises:
 filtering social media content users based on one or more filtering criteria, the filtering criteria comprising one or more of: i) social media user demographic information, ii) a content of the social media content items authored by the social media users, or iii) a time of authorship of the social media content items.