Topic and time based media affinity estimation
View Patent ↗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. A computer-executed method, comprising:
identifying a first population of social media users who are aligned with a topic;
accessing a repository comprising a plurality of time-based media events;
for each of the plurality of time-based media events:
identifying, by a computer processor, a second population of social media users who are aligned with the time-based media event, and
determining, by the computer processor and based on the first population and the second population, an affinity score associated with the time-based media event that is indicative of an affinity by social media users for both the topic and the time-based media event; and
ranking the time-based media events based on the affinity scores to identify a first time-based media event during which an advertisement associated with the topic should be aired.
2. The computer-executed method of claim 1 , further comprising determining a cost for airing the advertisement based on the affinity score associated with the first time-based media event.
3. The computer-executed method of claim 1 , further comprising:
determining that the advertisement aired during the first time-based media event;
identifying a third population of social media users who are aligned with the advertisement; and
determining based on the first population and the third population an affinity score associated with the advertisement that is indicative of an affinity by social media users for both the advertisement and the first time-based media event.
4. The computer-executed method of claim 3 , further comprising determining based on the affinity score associated with the advertisement that the advertisement should be aired during a second time-based media event included in the plurality of time-based media events.
5. The computer-executed method of claim 3 , further comprising:
identifying a fourth population of social media users who did not view the airing the advertisement;
determining based on the third population and the fourth population an inferred affinity score associated with the advertisement that is indicative of an affinity by social media users in the fourth population for the advertisement; and
determining based on the inferred affinity score whether the advertisement should be presented to the fourth population.
6. The computer-executed method of claim 1 , wherein, for a time-based media event, determining the second population who are aligned with the time-based media event comprises adding a social media user to the second population responsive to determining that at least one social media content item authored by the social user is aligned with the time-based media event.
7. The computer-executed method of claim 1 , wherein, for a time-based media event, determining an affinity score associated with the time-based media event comprises:
determining an intersection between the first population and the second population, each social media user in the intersection being aligned with both the topic and the time-based media event; and
computing the affinity score based on the social media users included in the intersection.
8. The computer-executed method of claim 7 , wherein computing the affinity score comprises assigning a weight to each social media user included in the intersection based on a level of alignment of the social media user with the topic and the time-based media event.
9. The computer-executed method of claim 8 , further comprising adjusting the weight assigned to each social media user based on how many social media content items the social media user has previously authored.
10. The computer-executed method of claim 7 , wherein computing the affinity score comprises:
determining an average probability of social media users in the intersection authoring social media content items related to the topic and the time-based media event; and
normalizing the average probability by an expected probability of any social media user authoring social media content items related to the topic and the time-based media event.
11. The computer-executed method of claim 1 , wherein, for a time-based media event, determining an affinity score associated with the time-based media event comprises:
filtering the first population based on a criteria to generate a filtered population;
determining an intersection between the filtered population and the second population; and
computing the affinity score based on the intersection.
12. The computer-executed method of claim 1 , comprising, responsive to identifying the first time-based media event, causing the advertisement to be aired during an airing of the first time-based media event.
13. The computer-executed method of claim 1 , comprising sending an advertisement to client devices associated with a population of social media users who are aligned with the time-based media event.
14. A system, comprising:
a computer processor; and
a computer-readable storage medium storing computer program engines configured to execute on the computer processor, the computer program engines comprising:
a population set aggregator configured to identify a first population of social media users who are aligned with a topic and a second population of social media users who are aligned with a time-based media event,
an affinity estimator configured to:
determine based on the first population and the second population an affinity score associated with the time-based media event that is indicative of an affinity by social media users for both the topic and the time-based media event, and
determine based on the affinity score that an advertisement associated with the topic should be aired during the airing of the time-based media event.
15. The system of claim 14 , wherein the affinity estimator is further configured to determine a cost for airing the advertisement based on the affinity score associated with the time-based media event.
16. The system of claim 14 , wherein
the population set aggregator is further configured to:
determine that the advertisement aired during the time-based media event, and identify a third population of social media users who are aligned with the advertisement; and
the affinity estimator is further configured to:
determine based on the first population and the third population an affinity score associated with the advertisement that is indicative of an affinity by social media users for both the advertisement and the time-based media event.
17. The system of claim 16 , wherein
the population set aggregator is further configured to:
identify a fourth population of social media users who did not view the airing the advertisement; and
the affinity estimator is further configured to:
determine based on the third population and the fourth population an inferred affinity score associated with the advertisement that is indicative of an affinity by social media users in the fourth population for the advertisement, and
determine based on the inferred affinity score whether the advertisement should be presented to the fourth population.
18. The system of claim 14 , wherein the population set aggregator identifies the second population by adding a social media user to the second population responsive to determining that at least one social media content item authored by the social user is aligned with the time-based media event.
19. The system of claim 14 , wherein the affinity estimator determines the affinity score by:
determining an intersection between the first population and the second population, each social media user in the intersection being aligned with both the topic and the time-based media event; and
computing the affinity score based on the social media users included in the intersection.
20. The system of claim 19 , wherein the affinity estimator computes the affinity score by assigning a weight to each social media user included in the intersection based on a level of alignment of the social media user with the topic and the time-based media event.
21. The system of claim 20 , wherein the affinity estimator computes the affinity score by adjusting the weight assigned to each social media user based on how many social media content items the social media user has previously authored.
22. The system of claim 19 , wherein the affinity estimator computes the affinity score by:
determining an average probability of social media users in the intersection authoring social media content items related to the topic and the time-based media event; and
normalizing the average probability by an expected probability of any social media user authoring social media content items related to the topic and the time-based media event.
23. The system of claim 14 , wherein the affinity estimator computes the affinity score by:
filtering the first population based on a criteria to generate a filtered population;
determining an intersection between the filtered population and the second population; and
computing the affinity score based on the intersection.
24. A computer readable medium storing instructions that, when executed by a computer processor, cause the computer processor to perform the steps of:
identifying a first population of social media users who are aligned with a topic;
accessing a repository comprising a plurality of time-based media events;
for each of the plurality of time-based media events:
identifying, by the computer processor, a second population of social media users who are aligned with the time-based media event, and
determining, by the computer processor, based on the first population and the second population an affinity score associated with the time-based media event that is indicative of an affinity by social media users for both the topic and the time-based media event; and
ranking the time-based media events based on the affinity scores to identify a first time-based media event during which an advertisement associated with the topic should be aired.