System and method for generating personal vocabulary from network data

• US 8,990,083 B1
• Filed: 09/30/2009
• Issued: 03/24/2015
• Est. Priority Date: 09/30/2009
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Gannu, Satish K., Malegaonkar, Ashutosh A., Mihailovici, Virgil N.
• Primary Examiner(s): JACKSON, JAKIEDA R
• Application Number:
  US12/571,404
• Time in Patent Office: 2,001 Days
• Field of Search: 704/240, 704238-239
• US Class Current: 704/240
• CPC Class Codes: G10L 15/06 Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice G10L15/14 takes precedence
• CPC Class Codes: G10L 2015/0633 using lexical or orthographic knowledge sources

Abstract

A method is provided in one example and includes receiving data propagating in a network environment, and identifying selected words within the data based on a whitelist. The whitelist includes a plurality of designated words to be tagged. The method further includes assigning a weight to the selected words based on at least one characteristic associated with the data, and associating the selected words to an individual. A resultant composite is generated for the selected words that are tagged. In more specific embodiments, the resultant composite is partitioned amongst a plurality of individuals associated with the data propagating in the network environment. A social graph can be generated that identifies a relationship between a selected individual and the plurality of individuals based on a plurality of words exchanged between the selected individual and the plurality of individuals.

Claims

1. A method, comprising:

receiving data propagating in a network environment at a streaming database feeder;

ignoring Joint Photographic Experts Group (JPEG) documents in the data;

updating tags for each user in the network environment using a user-sub stream created for the user by the streaming database feeder, wherein each user-sub stream includes at least a portion of the data propagating in the network environment, wherein the tags are words and phrases that are associated with each user, wherein the data includes documents and, for at least a portion of the documents in the data, each original document is copied to create an anonymous document and a document that contains selected words within the data based on a whitelist, wherein the whitelist includes a plurality of designated words to be tagged, wherein documents that include data in a blacklist are dropped, and wherein the anonymous documents contain a concept field and some of the data in the anonymous documents is selected for the whitelist, and wherein the document that contains selected words does not include the concept field;

assigning a weight to the selected words based on at least one characteristic associated with the data;

associating the selected words to an individual, wherein the weight for a selected word is higher if the individual propagates the data; and

generating a resultant composite of the selected words that are tagged.

2. The method of claim 1, further comprising:

partitioning the resultant composite amongst a plurality of individuals associated with the data propagating in the network environment; and

generating a social graph that identifies a relationship between a selected individual and the plurality of individuals based on a plurality of words exchanged between the selected individual and the plurality of individuals.

3. The method of claim 1, further comprising:

determining whether a threshold weight value associated with the selected words has been met, wherein if the threshold has been met, then the selected words are included in the resultant composite.

4. The method of claim 1, further comprising:

determining whether the data is sent or received by the individual, wherein a different weight value is assigned to the selected words based on the selected words being sent or received by the individual.

5. The method of claim 1, further comprising:

determining whether the selected words have been marked as private, wherein if the selected words are marked as private, then the selected words are not included in the resultant composite.

6. The method of claim 1, wherein the characteristic relates to a probability of the selected words appearing in a document that includes the data, and wherein individual words that have a lower probability of appearing in the document are assigned a higher weight value.

7. The method of claim 1, wherein the characteristic relates to a document type that includes the data.

8. The method of claim 1, wherein a plurality of anonymous documents are used to create a corporate vocabulary and a plurality of documents that contains selected words are used to create a personal vocabulary.

9. The method of claim 1, wherein the JPEG documents are excluded from the data.

10. Logic encoded in one or more non-transitory media that includes code for execution and when executed by a processor is operable to perform operations comprising:

receiving data propagating in a network environment at a streaming database feeder;

ignoring Joint Photographic Experts Group (JPEG) documents from the data;

updating tags for each user in the network environment using a user-sub stream created for the user by the streaming database feeder, wherein each user-sub stream includes at least a portion of the data propagating in the network environment, wherein the tags are words and phrases that are associated with each user, wherein the data includes documents and, for at least a portion of the documents in the data, each original document is copied to create an anonymous document and a document that contains selected words within the data based on a whitelist, wherein the whitelist includes a plurality of designated words to be tagged, wherein documents that include data in a blacklist are dropped, and wherein the anonymous documents contain a concept field and some of the data in the anonymous documents is selected for the whitelist, and wherein the document that contains selected words does not include the concept field;

assigning a weight to the selected words based on at least one characteristic associated with the data;

associating the selected words to an individual, wherein the weight for a selected word is higher if the individual propagates the data; and

generating a resultant composite of the selected words that are tagged.

11. The logic of claim 10, the processor being further operable to perform operations comprising:

partitioning the resultant composite amongst a plurality of individuals associated with the data propagating in the network environment; and

generating a social graph that identifies a relationship between a selected individual and the plurality of individuals based on a plurality of words exchanged between the selected individual and the plurality of individuals.

12. The logic of claim 10, the processor being further operable to perform operations comprising:

determining whether a threshold weight value associated with the selected words has been met, wherein if the threshold has been met, then the selected words are included in the resultant composite.

13. The logic of claim 10, the processor being further operable to perform operations comprising:

determining whether the data is sent or received by the individual, wherein a different weight value is assigned to the selected words based on the selected words being sent or received by the individual.

14. The logic of claim 10, the processor being further operable to perform operations comprising:

determining whether the selected words have been marked as private, wherein if the selected words are marked as private, then the selected words are not included in the resultant composite.

15. The logic of claim 10, wherein the characteristic relates to a probability of the selected words appearing in a document that includes the data, and wherein individual words that have a lower probability of appearing in the document are assigned a higher weight value.

16. The logic of claim 10, wherein the characteristic relates to a document type that includes the data.

17. An apparatus, comprising:

a memory element configured to store data;

a processor operable to execute instructions associated with the data;

a network sensor configured to interface with the memory element and the processor, the network sensor being configured to;

receive data propagating in a network environment at a streaming database feeder;

ignore Joint Photographic Experts Group (JPEG) documents from the data;

update tags for each user in the network environment using a user-sub stream created for the user by the streaming database feeder, wherein each user-sub stream includes at least a portion of the data propagating in the network environment, wherein the tags are words and phrases that are associated with each user, wherein the data includes documents and, for at least a portion of the documents in the data, each original document is copied to create an anonymous document and a document that contains selected words within the data based on a whitelist, wherein the whitelist includes a plurality of designated words to be tagged, wherein documents that include data in a blacklist are dropped, and wherein the anonymous documents contain a concept field and some of the data in the anonymous documents is selected for the whitelist, and wherein the document that contains selected words does not include the concept field; and

a weighting module configured to;

assign a weight to the selected words based on at least one characteristic associated with the data, wherein the selected words are associated to an individual and the weight for a selected word is higher if the individual propagates the data, and wherein a resultant composite of the selected words that are tagged is generated.

18. The apparatus of claim 17, further comprising:

a database configured to partition the resultant composite amongst a plurality of individuals associated with the data propagating in the network environment, and wherein the database includes a social graph that identifies a relationship between a selected individual and the plurality of individuals based on a plurality of words exchanged between the selected individual and the plurality of individuals.

19. The apparatus of claim 17, wherein the weighting module is further configured to determine whether a threshold weight value associated with the selected words has been met, wherein if the threshold has been met, then the selected words are included in the resultant composite.

20. The apparatus of claim 17, wherein a different weight value is assigned to the selected words based on the selected words being sent or received by the individual.