Methods, systems, and products for classifying content segments

• US 8,005,841 B1
• Filed: 04/28/2006
• Issued: 08/23/2011
• Est. Priority Date: 04/28/2006
• Status: Active Grant

• Current Assignee: Qurio Holdings Incorporated
• Original Assignee: Qurio Holdings Incorporated
• Inventors: Walsh, Richard J., Issa, Alfredo C.
• Primary Examiner(s): LIANG, VEI CHUNG
• Application Number:
  US11/414,021
• Time in Patent Office: 1,943 Days
• Field of Search: 707736-740
• US Class Current: 707/739
• CPC Class Codes: G06F 16/78 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually

Abstract

Methods, systems, and products are disclosed for classifying content segments. A set of annotations is received that occur within a segment of time-varying content. Each annotation is scored to each node in an ontology. The segment is classified based on at least one of the scores.

Claims

1. A device for classifying content comprising:

memory having a classification application stored thereon; and

a processor in communication with the memory, the processor configured to;

receive a set of annotations occurring within a segment of time-varying content;

score at least two annotations for the segment to each node in an ontology, wherein the score represents how each of the at least two annotations match each node in the ontology;

identify a sequence of annotations that have a nodal score for the same node that exceeds a minimum threshold score;

create a semantic segment from the sequence of annotations;

merge the semantic segment with other semantic segments that satisfy a minimum threshold score;

sum the scores of the at least two annotations for each node and determine a semantic distance between each node of the ontology for the segment, thereby determining a total score for each node based on the summed score and the semantic distance between each node where the total score is increased based on the semantic distance; and

classify the segment based on the total score for each node.

2. The device according to claim 1, wherein the processor is further configured to receive metadata associated with each annotation that references at least one of:

i) an ontological node within the ontology;

ii) a strength of an association between an annotation and the ontological node; and

iii) a communications address identifying a location of the ontology.

3. The device according to claim 1, wherein the processor is further configured to compare each score to a minimum threshold score, and classify the segment based on a score greater than or equal to the minimum threshold score.

4. The device according to claim 1, wherein the processor is further configured to calculate a nodal score for each node and compare each nodal score to a minimum threshold score, and classify the segment based on a nodal score greater than or equal to the minimum threshold score.

5. The device according to claim 1, wherein the processor is further configured to calculate a nodal score for each node and modify each nodal score by the semantic distance between at least two nodes.

6. The device according to claim 1, wherein the processor is further configured to calculate a nodal score for each node and collapse at least one lower scoring node into a higher scoring node.

7. The device according to claim 1, wherein the processor is further configured to store the classification in memory.

8. The device according to claim 1, wherein the processor is further configured to compare nodal scores for adjacent annotations.

9. The device according to claim 1, wherein the processor is further configured to exclude an annotation from the sequence that exceeds a threshold time.

10. The device according to claim 1, wherein the relationship between each node of the ontology for the segment includes a nodal distance between each node.

11. A system, comprising:

a classification application stored in memory; and

a processor communicating with the memory and with the classification application, the system configured to;

receive a set of annotations occurring within a segment of time-varying content;

score at least two annotations for the segment to each node in an ontology, wherein the score represents how each of the at least two annotations match each node in the ontology;

identify a sequence of annotations that have a nodal score for the same node that exceeds a minimum threshold score;

create a semantic segment from the sequence of annotations;

merge the semantic segment with other semantic segments that satisfy a minimum threshold score;

sum the scores of the at least two annotations for each node and determine a semantic distance between each node of the ontology for the segment, thereby determining a total score for each node based on the summed score and the semantic distance between each node where the total score is increased based on the semantic distance; and

classify the segment based on the total score for each node.

12. The system according to claim 11, wherein the processor compares each score to a minimum threshold score, and classifies the segment based on a score greater than or equal to the minimum threshold score.

13. The system according to claim 11, wherein the processor is configured to:

i) calculates a nodal score for each node,

ii) compares each nodal score to a minimum threshold score, and

iii) classifies the segment based on a nodal score greater than or equal to the minimum threshold score.

14. The system according to claim 11, wherein the relationship between each node of the ontology for the segment includes a nodal distance between each node.

15. A computer program product comprising a computer-readable medium storing instructions for classifying content segments, the instructions instructing a computer to perform the following steps:

receive a set of annotations occurring within a segment of time-varying content;

score at least two annotations for the segment to each node in an ontology, wherein the score represents how each of the at least two annotations match each node in the ontology;

identify a sequence of annotations that have a nodal score for the same node that exceeds a minimum threshold score;

create a semantic segment from the sequence of annotations;

merge the semantic segment with other semantic segments that satisfy a minimum threshold score;

sum the scores of the at least two annotations for each node and determine a semantic distance between each node of the ontology for the segment, thereby determining a total score for each node based on the summed score and the semantic distance between each node where the total score is increased based on the semantic distance; and

classify the segment based on the total score for each node.

16. The computer program product according to claim 15, wherein the relationship between each node of the ontology for the segment includes a nodal distance between each node.