Granted Patent B1

US 7,899,871 · App. 11/838,738 · Granted Mar 1, 2011

Abstract

Claims (81)

1. A method for processing e-mails, the method comprising:

receiving, at one or more computer systems, a plurality of e-mails obtained from an e-mail repository;

storing information representing each e-mail in the plurality of e-mails in a storage device associated with the one or more computer systems, stored information representing a given e-mail including at least one of content of the given e-mail and message attribute data of the given e-mail;

for each e-mail in the plurality of e-mails,

determining, with one or more processors associated with the one or more computer systems, a set of noun phrases included in the e-mail based on stored information representing the e-mail, and

generating, with the one or more processors associated with the one or more computer systems, a feature representation for the e-mail based on the determined set of noun phrases included in the e-mail, a feature representation for a given e-mail scoring selected noun phrases in a set of noun phrases determined to be included in the given e-mail;

generating, with the one or more processors associated with the one or more computer systems, a set of topics based on the feature representation generated for each e-mail in the plurality of emails; and

storing each topic in the set of topics in a storage device associated with the one or more computer systems in association with at least some of the plurality of e-mails.

2. The method of claim 1 wherein determining the set of noun phrases included in the e-mail comprises:

determining sentence structure associated with the e-mail; and

determining parts of speech associated with the e-mail, wherein the parts of speech include the set of noun phrases.

3. The method of claim 1 wherein generating the set of topics based on the feature representation generated for each e-mail in the plurality of e-mails comprises:

determining a centroid associated with a cluster of feature representations generated for one or more e-mails in the plurality of e-mails; and

generating a topic based on the centroid, the generated topic represented by a predetermined number of noun phrases common to the feature representations generated for the one or more e-mails.

4. The method of claim 1 further comprising:

determining a name associated with a topic in the set of topic in the set of topics based on the feature representation of one or more e-mails in the plurality of e-mails; and

associating the name with the topic.

5. The method of claim 1 further comprising:

identifying a topic in the set of topics as relevant to one or more search terms;

identifying one or more e-mails in the plurality of e-mails that are associated with the identified topic; and

generating information configured for displaying the identified one or more e-mails together with the one or more search terms.

6. The method of claim 1 further comprising:

ranking each topic in the set of topics; and

generating information configured for displaying selected topics in the set of topics according to the rank of the selected topics.

7. The method of claim 1 further comprising:

determining a hierarchy between a first topic in the set of topics and a second topic in the set of topics; and

generating information configured for displaying selected e-mails in the plurality of emails that are associated with the first topic based on the hierarchy between the first topic and the second topic.

8. A non-transitory computer-readable medium storing a set of code modules which when executed by a processor of a computer system cause the processor to process e-mails, the computer-readable medium comprising:

code for receiving a plurality of e-mails obtained from an e-mail repository;

code for storing information representing each e-mail in the plurality of e-mails, stored information representing a given e-mail including at least one of content of the given e-mail and message attribute data of the given e-mail;

code for determining, for each e-mail in the plurality of e-mails, a set of noun phrases included in an e-mail based on the stored information representing the e-mail;

code for generating, for each e-mail in the plurality of e-mails, a feature representation for an e-mail based on a determined set of noun phrases associated with included in the e-mail, a feature representation for a given e-mail scoring selected noun phrases in a determined set of noun phrases included in the given e-mail;

code for generating a set of topics based on a feature representation generated for each e-mail in the plurality of e-mails; and

code for storing each topic in the set of topics in association with at least some of the plurality of e-mails.

9. The computer-readable medium of claim 8 wherein the code for generating, for each e-mail in the plurality of e-mails, a feature representation for an e-mail based on a determined set of noun phrases included in the e-mail comprises:

code for determining sentence structure associated with the e-mail; and

code for determining parts of speech associated with the e-mail, wherein the parts of speech include the set of noun phrases.

10. The computer-readable medium of claim 8 wherein the code for generating the set of topics based on a feature representation generated for each e-mail in the plurality of e-mails comprises:

code for determining a centroid associated with a cluster of feature representations generated for one or more e-mails in the plurality of e-mails; and

code for generating a topic in the set of topics based on the centroid, the generated topic represented by a predetermined number of noun phrases common to the feature representations generated for the one or more e-mails.

11. The computer-readable medium of claim 8 further comprising:

code for determining a name associated with a topic in the set of topics based on the feature representation of one or more e-mails in the plurality of e-mails; and

code for associating the name with the topic.

12. The computer-readable medium of claim 8 further comprising:

code for identifying a topic in the set of topics as relevant to one or more search terms;

code for identifying one or more e-mails in the plurality of e-mails that are associated with the identified topic; and

code for displaying the identified one or more e-mails in response to the one or more search terms.

13. The computer-readable medium of claim 8 further comprising:

code for ranking each topic in the set of topics; and

code for displaying selected topics in the set of topics according to the rank of the selected topics.

14. The computer-readable medium of claim 8 further comprising:

code for determining a hierarchy between a first topic in the set of topics and a second topic in the set of topics; and

code for displaying selected e-mails in the plurality of emails that are associated with the first topic based on the hierarchy between the first topic and the second topic.

15. A system for processing e-mails, the system comprising:

a processor; and

a memory coupled to the processor, the memory configured to store a set of instructions which when executed by the processor configure the processor to:

receive a plurality of e-mails obtained from an e-mail repository;

store information representing each e-mail in the plurality of e-mails, stored information representing a given e-mail including at least one of content of the given e-mail and message attribute data of the given e-mail;

for each e-mail in the plurality of e-mails,

determine a set of noun phrases included in the e-mail based on stored information representing the e-mail, and

generate a feature representation for the e-mail based on the determined set of noun phrases included in the e-mail, a feature representation for a given e-mail scoring selected noun phrases in a set of noun phrases determined to be included in the given e-mail;

generate a set of topics based on the feature representation generated for each e-mail in the plurality of e-mails; and

store each topic in the set of topics in association with at least some of the plurality of e-mails.

16. The system of claim 15 wherein, to determine the set of noun phrases included in the e-mail based on the stored information representing the e-mail, the processor is configured to:

determine sentence structure associated with the e-mail; and

determine parts of speech associated with the e-mail, wherein the parts of speech include the set of noun phrases.

17. The system of claim 15 wherein, to generate the set of topics based on the feature representation generated for each e-mail in the plurality of e-mails, the processor is configured to:

determine a centroid associated with a cluster of feature representations generated for one or more e-mails in the plurality of e-mails; and

determine a topic based on the centroid, the generated topic represented by a predetermined number of noun phrases common to the feature representations generated for the one or more e-mails.

18. The system of claim 15 wherein the processor is further configured to:

determine a name associated with a topic in the set of topics based on the feature representation of one or more e-mails in the plurality of e-mails; and

associate the name with the topic.

19. The system of claim 15 wherein the processor is further configured to:

identify a topic in the set of topics as relevant to one or more search terms; and

identify one or more e-mails in the plurality of e-mails that are associated with the identified topic.

20. The system of claim 15 wherein the processor is further configured to:

rank each topic in the set of topics; and

generating information configured for displaying selected topics in the set of topics according to the rank of the selected topics.

21. The system of claim 15 wherein the processor is further configured to:

determine a hierarchy between a first topic in the set of topics and a second topic in the set of topics; and

generating information configured for displaying selected e-mails in the plurality of emails that are associated with the first topic based on the hierarchy between the first topic and the second topic.

Assignments (9)