System, method, and computer program for segmenting a content stream

• US 8,994,311 B1
• Filed: 10/05/2010
• Issued: 03/31/2015
• Est. Priority Date: 05/14/2010
• Status: Active Grant

• Current Assignee: Amdocs Development Limited (Amdocs Ltd.), Amdocs Software Systems Limited (Amdocs Ltd.)
• Original Assignee: Amdocs Software Systems Limited (Amdocs Ltd.)
• Inventors: Balfe, Evelyn, Lynch, James, Smyth, Barry J., Cotter, Paul
• Primary Examiner(s): Alam, Hosain
• Assistant Examiner(s): Hosseini, Reza
• Application Number:
  US12/898,641
• Time in Patent Office: 1,638 Days
• Field of Search: 707/614, 707/618, 707/673, 707/696, 707/706, 707/707, 707/711, 707/715, 707/722, 707/723, 707/733, 707/734, 707/741, 707/749, 707/753, 707/754, 707759-770, 707/780, 707/784
• US Class Current: 318/565
• CPC Class Codes: G06F 16/7867 using information manually generated, e.g. tags, keywords, comments, title and artist information, manually generated time, location and usage information, user ratings
• CPC Class Codes: G06F 16/958 Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking

Abstract

A system, method, and computer program product are provided for segmenting a content stream. In use, a content stream associated with an event is identified. Additionally, user generated content associated with the content stream is identified. Further, a timing of one or more key events within the content stream is determined, based on the user generated content. Further still, the one or more key events are identified, based on the user generated content. Also, the content stream is segmented, utilizing the one or more key events and the timing of the one or more key events. In addition, a subset of the user generated content and a subset of the one or more key events is associated with a segment of the content stream. Furthermore, a search index associated with the segment of the content stream is generated, utilizing the subset of the user generated content and the subset of the one or more key events. Further still, a search query is received from a user. Also, search results are provided to the user in response to the search query, utilizing the search index associated with the segment of the content stream.

Claims

1. A computer program embodied on a non-transitory computer readable medium, comprising:

computer code for identifying a content stream associated with an event, wherein the content stream includes one or more of video and audio that is streamed to a user;

computer code for identifying a plurality of instances of user generated content separate from and associated with the content stream, wherein the plurality of instances of user generated content include a plurality of instances of textual commentary about the content stream that are published via social media synchronously with at least one of the event and the streaming of the content to the user and include one or more of a web log entry about the content stream, a tweet about the content stream, a short message service (SMS) message about the content stream, a phone call about the content stream, social commentary about the content stream, and automatically generated content;

computer code for determining a timing of a plurality of key events within the content stream, based on the plurality of instances of user generated content, where the timing of each of the plurality of key events includes a time of occurrence of each of the plurality of key events and is determined by;

synchronizing the content stream with the plurality of instances of user generated content with respect to time,

calculating a number of the plurality of instances of the user generated content that fall within the predetermined timing, and

identifying the predetermined timing as the timing of the key event when the number of the plurality of instances of the user generated content calculated for the timing meets a predetermined number;

computer code for identifying the plurality of key events within the content stream, including for each of the plurality of key events;

identifying a plurality of descriptions within the user generated content,

identifying a timing when each of the plurality of descriptions within the user generated content was generated,

correlating the timing of one or more of the plurality of descriptions with the determined timing of the key event, and

labeling the key event with the one or more of the plurality of descriptions correlating with the key event;

computer code for segmenting the content stream into a plurality of segments, such that each segment contains one or more of the identified plurality of key events;

computer code for, for each segment of the content stream;

associating a subset of the user generated content and a subset of the plurality of key events with the segment of the content stream, and

generating a search index associated with the segment of the content stream, utilizing the associated subset of the user generated content and the associated subset of the plurality of key events;

computer code for receiving a search query from another user;

computer code for searching a plurality of search indexes including the plurality of generated search indexes based on the search query, by comparing the search query to the plurality of search indexes; and

computer code for providing search results to the other user based on the searching;

wherein the computer program is operable such that when the search query matches at least a portion of a particular one of the plurality of generated search indexes associated with a particular one of the segments of the content stream, the search results include the particular one of the segments of the content stream.

2. The computer program of claim 1, wherein the content includes one or more of video content, audio content, and text content.

3. The computer program of claim 1, wherein the computer program is operable such that the user generated content directly or indirectly addresses one or more elements of the content stream.

4. The computer program of claim 1, wherein the plurality of key events include a plurality of points in the content stream that are determined to be significant.

5. The computer program of claim 1, wherein the computer program is operable such that a traffic analyzer is used to determine the frequency of publication of new instances of the user generated content.

6. The computer program of claim 1, wherein the computer code for identifying the plurality of key events further comprises computer code for identifying the descriptions within the user generated content, using one or more machine learning algorithms.

7. The computer program of claim 1, wherein the descriptions include one or more keywords, tags, or topics.

8. The computer program of claim 1, wherein segmenting the content stream includes marking the plurality of key events within the content stream.

9. The computer program of claim 1, wherein the computer program is operable such that the search results are tailored according to one or more elements associated with the other user.

10. The computer program of claim 1, wherein the computer program is operable such that the search results are ranked based on one or more portions of the user generated content being associated with the other user.

11. The computer program of claim 1, further comprising computer code for providing a personal summary of the plurality of key events to the other user, based on a profile of the other user.

17. The computer program of claim 1, wherein the computer code for associating the subset of the user generated content and the subset of the plurality of key events with the segment of the content stream includes computer code for associating the segment of the content stream with the subset of the user generated content that was generated during a time the segment of the content stream was shown and the subset of the plurality of key events where the timing of each of the key events in the subset occurs during the time the segment of the content stream was shown.

18. The computer program of claim 1, wherein the computer code for generating the search index associated with the segment of the content stream, utilizing the associated subset of the user generated content and the associated subset of the plurality of key events includes computer code for generating the search index associated with the segment of the content stream by:

identifying a plurality of terms associated with the segment that occur within the associated subset of the user generated content generated during the time the segment was shown,

determining a term frequency score for each of the identified plurality of terms associated with the segment,

determining one or more key terms for the segment of the content stream by determining frequency scores for one or more of the identified plurality of terms associated with the segment that are greater than frequency scores for the one or more of the identified plurality of terms associated with one or more other segments, and

including the one or more key terms in the search index associated with the segment.

12. A method, comprising:

identifying a content stream associated with an event, wherein the content stream includes one or more of video and audio that is streamed to a user;

identifying a plurality of instances of user generated content separate from and associated with the content stream, wherein the plurality of instances of user generated content include a plurality of instances of textual commentary about the content stream that are published via social media synchronously with at least one of the event and the streaming of the content to the user and include one or more of a web log entry about the content stream, a tweet about the content stream, a short message service (SMS) message about the content stream, a phone call about the content stream, social commentary about the content stream, and automatically generated content;

determining a timing of a plurality of key events within the content stream, based on the plurality of instances of user generated content, where the timing of each of the plurality of key events is determined by;

synchronizing the content stream with the plurality of instances of user generated content with respect to time,

calculating a number of the plurality of instances of the user generated content that fall within the predetermined timing, and

identifying the predetermined timing as the timing of the key event when the number of the plurality of instances of the user generated content calculated for the timing meets a predetermined number;

identifying the plurality of key events within the content stream, including for each of the plurality of key events;

identifying a plurality of descriptions within the user generated content,

identifying a timing when each of the plurality of descriptions within the user generated content was generated,

correlating the timing of one or more of the plurality of descriptions with the determined timing of the key event, and

labeling the key event with the one or more of the plurality of descriptions correlating with the key event;

segmenting the content stream into a plurality of segments, such that each segment contains one or more of the identified plurality of key events;

for each segment of the content stream;

associating a subset of the user generated content and a subset of the plurality of key events with the segment of the content stream, and

generating a search index associated with the segment of the content stream, utilizing the associated subset of the user generated content and the associated subset of the plurality of key events;

receiving a search query from another user;

searching a plurality of search indexes including the plurality of generated search indexes based on the search query, by comparing the search query to the plurality of search indexes; and

providing search results to the other user based on the searching;

wherein when the search query matches at least a portion of a particular one of the plurality of generated search indexes associated with a particular one of the segments of the content stream, the search results include the particular one of the segments of the content stream.

15. The method of claim 12, further including providing, to a third party, a chronological characterization of the plurality of key events based on the user generated content.

16. The method of claim 12, wherein the timestamp associated with the subset of the user generated content indicates one of a time at which the subset of the user generated content was created, posted, and sent.

13. A system, comprising:

a processor for;

identifying a content stream associated with an event, wherein the content stream includes one or more of video and audio that is streamed to a user;

identifying a plurality of instances of user generated content separate from and associated with the content stream, wherein the plurality of instances of user generated content include a plurality of instances of textual commentary about the content stream that are published via social media synchronously with at least one of the event and the streaming of the content to the user and include one or more of a web log entry about the content stream, a tweet about the content stream, a short message service (SMS) message about the content stream, a phone call about the content stream, social commentary about the content stream, and automatically generated content;

determining a timing of a plurality of key events within the content stream, based on the plurality of instances of user generated content, where the timing of each of the plurality of key events is determined by;

synchronizing the content stream with the plurality of instances of user generated content with respect to time,

calculating a number of the plurality of instances of the user generated content that fall within the predetermined timing, and

identifying the predetermined timing as the timing of the key event when the number of the plurality of instances of the user generated content calculated for the timing meets a predetermined number;

identifying the plurality of key events within the content stream, including for each of the plurality of key events;

identifying a plurality of descriptions within the user generated content,

identifying a timing when each of the plurality of descriptions within the user generated content was generated,

correlating the timing of one or more of the plurality of descriptions with the determined timing of the key event, and

labeling the key event with the one or more of the plurality of descriptions correlating with the key event;

segmenting the content stream into a plurality of segments, such that each segment contains one or more of the identified plurality of key events;

for each segment of the content stream;

associating a subset of the user generated content and a subset of the plurality of key events with the segment of the content stream, and

generating a search index associated with the segment of the content stream, utilizing the associated subset of the user generated content and the associated subset of the plurality of key events;

receiving a search query from another user;

searching a plurality of search indexes including the plurality of generated search indexes based on the search query, by comparing the search query to the plurality of search indexes; and

providing search results to the other user based on the searching;

wherein when the search query matches at least a portion of a particular one of the plurality of generated search indexes associated with a particular one of the segments of the content stream, the search results include the particular one of the segments of the content stream.

14. The system of claim 13, wherein the processor is coupled to memory via a bus.