SYSTEM AND METHOD OF SEMANTIC CORRELATION OF RICH CONTENT

• US 20070106651A1
• Filed: 12/26/2006
• Published: 05/10/2007
• Est. Priority Date: 07/13/2000
• Status: Active Grant

• Current Assignee: RPX Corporation
• Original Assignee: Novell Incorporated (Open Text Corporation)
• Inventors: Isaacson, Scott, Carter, Steve
• Granted Patent: US 7,672,952 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 18/2413 based on distances to training or reference patterns
• CPC Class Codes: H04N 21/4394 involving operations for analysing the audio stream, e.g. detecting features or characteristics in audio streams arrangements characterised by components specially adapted for monitoring, identification or recognition of audio in broadcast systems H04H60/58
• CPC Class Codes: H04N 21/4542 Blocking scenes or portions of the received content, e.g. censoring scenes
• CPC Class Codes: H04N 21/466 Learning process for intelligent management, e.g. learning user preferences for recommending movies services using the results of monitoring in broadcast systems H04H60/61
• CPC Class Codes: H04N 21/4668 for recommending content, e.g. movies
• CPC Class Codes: H04N 21/8586 by using a URL processing chained hypermedia data for information retrieval G06F16/94; information retrieval from the Internet by using URLs G06F16/955; URL in broadcast information H04H20/93; Web-based protocols H04L67/02
• CPC Class Codes: H04N 7/163 by receiver means only

Abstract

Content of interest to a user is identified. A first semantic abstract is generated for the content of interest using a state vector representation of the content of interest. The first semantic abstract is compared to a second semantic abstract representing another prospective content. If the first semantic abstract is within a threshold distance of the second semantic abstract, then the prospective content is recommended to the user.

Claims

1. A system to recommend content to a user, comprising:

a content of interest to the user;

a first semantic abstract representing the content of interest;

a second semantic abstract representing a second content;

a semantic abstract comparer to compare the first semantic abstract to the second semantic abstract; and

a content recommender to recommend the second content if the first semantic abstract is within a threshold distance of the second semantic abstract.

2. A system according to claim 1, further comprising:

a content receiver to receive the content of interest;

a content presenter to present the content of interest to the user; and

a content of interest identifier to identify the content of interest after receiving a user request for the content of interest.

3. A system according to claim 1, further comprising:

a content monitor to monitor the content of interest;

a phrase extractor to extract phrases in the content of interest; and

a semantic abstract generator to generate the first semantic abstract for the content of interest using the phrases.

4. A system according to claim 3, wherein the semantic abstract generator includes:

a state vector constructor to construct state vectors in a topological vector space for each phrase extracted by the phrase extractor, the state vectors measuring how concretely each phrase extracted by the phrase extractor is represented in each chain in a basis and a dictionary, the dictionary including a directed set of concepts including a maximal element and at least one chain from the maximal element to every concept in the directed set, the basis including a subset of chains in the directed set; and

a collector to collect the state vectors into the first semantic abstract for the content of interest.

5. A system according to claim 1, further comprising an interface to receive a level of interest for the content of interest.

6. A system according to claim 1, further comprising a content library to store at least one of the content of interest and the second content.

7. A system according to claim 6, wherein the content recommender includes a link to the second content.

8. A system according to claim 1, wherein the content recommender includes an alert to notify the user of a future presentation of the second content.

9. A system according to claim 1, further comprising a preferred semantic abstract storage to store the first semantic abstract for the content of interest.

10. A system according to claim 9, wherein the preferred semantic abstract storage includes a preferred semantic abstract storage for a community of users.

11. A computer-implemented method for identifying and recommending content to a user, comprising:

identifying a first semantic abstract corresponding to a content of interest;

identifying a second semantic abstract corresponding to a second content;

measuring a distance between the first semantic abstract to the second semantic abstract; and

recommending the second content if the first semantic abstract is within a threshold distance of the second semantic abstract.

12. A method according to claim 11, further comprising:

receiving the content of interest;

identifying the content of interest after receiving a user request for the content of interest; and

presenting the content of interest.

13. A method according to claim 11, further comprising:

monitoring the content of interest;

extracting dominant phrases from the content of interest; and

generating the semantic abstract for the content of interest using the dominant phrases.

14. A method according to claim 13, wherein generating the semantic abstract for the content of interest includes:

identifying a directed set of concepts as a dictionary, the directed set including a maximal element and at least one concept, and at lest one chain from the maximal element to every concept;

selecting a subset of the chains to form a basis for the dictionary;

measuring how concretely the dominant phrases are represented in each chain in the basis and the dictionary;

constructing at least one state vector in the topological vector space for each dominant phrase in using the measures of how concretely each dominant phrase is represented in each chain in the dictionary and the basis; and

collecting the state vectors into the semantic abstract for the content of interest.

15. A method according to claim 14, wherein measuring a distance includes measuring a Hausdorff distance between the first semantic abstract and the second semantic abstract.

16. A method according to claim 11, wherein identifying a second semantic abstract corresponding to a second content includes:

identifying dominant words in the first content; and

searching for the second content using the dominant words.

17. A method according to claim 16, further comprising:

generating the second semantic abstract for the second content; and

confirming the second content is similar to the content of interest if the first semantic abstract is within a threshold distance of the second semantic abstract.

18. A method according to claim 11, wherein identifying a first semantic abstract includes identifying the first semantic abstract in a semantic abstract storage.

19. A method according to claim 11, wherein recommending the second content includes presenting a link to the second content.

20. A method according to claim 11, wherein recommending the second content includes alerting the user about a future presentation of the second content.

21. An article, comprising a storage medium, said storage medium having stored thereon instructions, that, when executed by a machine, result in:

identifying a first semantic abstract corresponding to a content of interest;

identifying a second semantic abstract corresponding to a second content;

comparing the first semantic abstract to the second semantic abstract; and

recommending the second content if the first semantic abstract is within a threshold distance of the second semantic abstract.

22. An article according to claim 21, further comprising:

receiving the content of interest;

identifying the content of interest after receiving a user request for the content of interest; and

presenting the content of interest,

23. An article according to claim 21, further comprising:

monitoring the content of interest;

extracting dominant phrases from the content of interest; and

generating the semantic abstract for the content of interest using the dominant phrases.

24. An article according to claim 23, wherein generating the semantic abstract for the content of interest includes:

identifying a directed set of concepts as a dictionary, the directed set including a maximal element and at least one concept, and at lest one chain from the maximal element to every concept;

selecting a subset of the chains to form a basis for the dictionary;

measuring how concretely the dominant phrases are represented in each chain in the basis and the dictionary;

constructing at least one state vector in the topological vector space for each dominant phrase in using the measures of how concretely each dominant phrase is represented in each chain in the dictionary and the basis; and

collecting the state vectors into the semantic abstract for the content of interest.

25. An article according to claim 24, wherein measuring a distance includes measuring a Hausdorff distance between the first semantic abstract and the second semantic abstract.

26. An article according to claim 21, wherein identifying a second semantic abstract corresponding to a second content includes:

identifying dominant words in the first content; and

searching for the second content using the dominant words.

27. A method according to claim 26, further comprising:

generating the second semantic abstract for the second content; and

confirming the second content is similar to the content of interest if the first semantic abstract is within a threshold distance of the second semantic abstract.