Displaying estimated social interest in time-based media
Social media content items are mapped to relevant time-based media events. These mappings may be used as the basis for multiple applications, such as ranking of search results for time-based media, automatic recommendations for time-based media, prediction of audience interest for media purchasing/planning, and estimating social interest in the time-based media. Social interest in time-based media (e.g., video and audio streams and recordings) segments is estimated through a process of data ingestion and integration. The estimation process determines social interest in specific events represented as segments in time-based media, such as particular plays in a sporting event, scenes in a television show, or advertisements in an advertising block. The resulting estimates of social interest also can be graphically displayed.
1. A computer-executed method for displaying social interest in television programs, the method comprising:
storing a plurality of social media content items received from an external social networking system;
selecting a plurality of television programs, each television program associated with one of a series of chronological time segments of television;
for each selected television program:
determining that a subset of the social media content items is relevant to the television program,
storing the subset of the social media content items relevant to the television program, and
determining a level of social interest in the television program based on the subset of the social media content items determined to be relevant to the television program;
for each of the time segments, determining a level of social interest in the time segment based upon an aggregate level of social interest in the television programs associated with the time segment; and
graphically displaying at least one time segment and the determined level of social interest for each displayed time segment.
2. The computer-executed method of claim 1 , wherein graphically displaying at least one time segment comprises displaying different television program types such that the different television program types each are visually distinct.
3. The computer-executed method of claim 1 , further comprising filtering the time segments according to a search term, wherein graphically displaying at least one time segment comprises displaying only a subset of the series of chronological time segments corresponding to the search term.
4. The computer-executed method of claim 1 , wherein graphically displaying at least one time segment comprises:
determining a first portion of the time segment corresponding to a positive sentiment;
determining a second portion of the time segment correspond to a negative sentiment; and
displaying the time segment with the first portion and the second portion visually distinguished from each other.
5. The computer-executed method of claim 1 , wherein determining the level of social interest in the television program comprises:
extracting event features from annotations associated with the television program;
for each of the plurality of social media content items:
extracting a social media feature from the social media content item;
determining, based on the social media feature and a subset of the event features, a confidence score indicative of the probability that the social media content item is relevant to the television program; and
mapping the television program to the subset of social media content items based on the confidence scores of the subset of social media content items.
6. The computer-executed method of claim 1 , wherein graphically displaying at least one time segment comprises:
graphically displaying a social interest heat map area comprising a social interest heat map showing the level of social interest for the time segment;
graphically displaying a media display area comprising a television program selected from the social interest heat map, wherein the media display area is visually distinguished from and concurrently displayed with the social interest heat map area; and
graphically displaying a social media display area comprising social media content items for the selected television program, wherein the social media display area is visually distinguished from and concurrently displayed with the social interest heat map area and the media display area.
7. The computer-executed method of claim 6 , wherein the social interest heat map area displays different event types such that the different event types each are visually distinct.
8. The computer-executed method of claim 6 , further comprising graphically displaying a field for filtering the time segments according to a search term, wherein the social interest heat map area displays only a subset of the time segments corresponding to the search term.
9. The computer-executed method of claim 1 , wherein graphically displaying at least one time segment comprises displaying a first portion of the time segment corresponding to positive sentiment that is visually distinguished from a second portion of the time segment corresponding to negative sentiment.
10. The computer-executed method of claim 1 , wherein the social media content items as received from the social networking system do not each include metadata directly and individually associating them with individual television programs.
11. The computer-executed method of claim 1 , wherein at least one of the social media content items does not include information explicitly associated with a television program.
12. The computer-executed method of claim 1 , wherein at least one of the social media content items does not contain metadata that uniquely identifies a television program.
13. The computer-executed method of claim 1 , wherein each time segment is associated with a different subset of the plurality of television programs.
14. The computer-executed method of claim 1 , wherein at least one of the social media content items comprises text authored by a user of the social networking system.
15. A system for displaying social interest in television programs, the system comprising:
means for storing a plurality of social media content items received from an external social networking system;
means for selecting a plurality of television programs, each television program associated with one of a series of chronological time segments of television;
means for determining, for each selected television program, that a subset of the social media content items is relevant to the television program, and storing the subset of the social media content items relevant to the television program;
means for determining, for each selected television program, a level of social interest in the television program based on the subset of the social media content items determined to be relevant to the television program;
means for determining for each of the time segments a level of social interest in the time segment based upon an aggregate level of social interest in the television programs associated with the time segment; and
means for graphically displaying at least one time segment and the determined level of social interest for each displayed time segment.
16. The system of claim 15 , wherein determining the level of social interest in the television program comprises:
extracting event features from annotations associated with the television program;
for each of the plurality of social media content items:
extracting a social media feature from the social media content item; and
determining, based on the social media feature and a subset of the event features, a confidence score indicative of the probability that the social media content item is relevant to the television program; and
mapping the television program to the subset of social media content items based on the confidence scores of the subset of social media content items.
17. A system for displaying social interest in television programs, the system comprising:
a computer processor;
a social media content item store configured to store a plurality of social media content items received from an external social networking system;
a multimedia store configured to store a plurality of selected television programs, each television program associated with one of a series of chronological time segments of television;
a social interest estimator executing on the computer processor and configured to:
for each selected television program:
determine that a subset of the social media content items is relevant to the television program,
store the subset of the social media content items relevant to the television program, and
determine a level of social interest in the television program based on the subset of the social media content items determined to be relevant to the television program, and
determine for each of the time segments a level of social interest in the time segment based upon an aggregate level of social interest in the television programs associated with the time segment; and
a user interface engine configured to graphically display at least one time segment and the determined level of social interest for each displayed time segment.
18. The system of claim 17 , wherein graphical displaying at least one time segment comprises displaying different television program types such that the different television program types each are visually distinct.
19. The system of claim 17 , further comprising a domain ontology engine configured to filter the time segments according to a search term, wherein graphical displaying at least one time segment comprises displaying only a subset of the series of chronological time segments corresponding to the search term.