Classification of electronic messages based on content

• US 8,140,540 B2
• Filed: 03/16/2009
• Issued: 03/20/2012
• Est. Priority Date: 03/16/2009
• Status: Active Grant

• Current Assignee: Terrace Licensing LLC (IP Edge LLC)
• Original Assignee: International Business Machines Corporation
• Inventors: Chitiveli, Srinivas V., Emanuel, Barton W., Holt, Alexander W., Moran, Michael E.
• Primary Examiner(s): Nguyen, Phong
• Application Number:
  US12/404,716
• Publication Number:
  US 20100235367A1
• Time in Patent Office: 1,100 Days
• Field of Search: 707/748, 707/749, 707/752, 707/771
• US Class Current: 707/748
• CPC Class Codes: G06F 16/353 into predefined classes

Abstract

Classifying electronic mail (e-mail) based on content and predefined categories. Content of a received e-mail may be analyzed to determine one of a plurality of predefined categories into which the e-mail is classified. A relevancy score may also be calculated to indicate the strength of correlation between the e-mail and the category. A user may be allowed to sort e-mails in an e-mail box based on the category names and/or relevancy scores.

Claims

1. A computer-implemented method for classifying an electronic message, comprising:

configuring one or more computer processors with an electronic message program to perform an operation comprising;

providing access to a predefined list of categories, wherein each of the categories has a predefined association with a respective one or more key words, and wherein each of the one or more key words has an associated weight value reflecting a correlation between the respective key word and the category;

for each category in the predefined list of categories, calculating, by operation of one or more computer processors, a relevancy score based on the one or more key words found in a body of the electronic message, wherein the relevancy score reflects a correlation between the electronic message and the category;

associating the electronic message with a category from the predefined list of categories having a strongest correlation with the electronic message based on the calculated relevancy scores; and

responsive to detecting a user action associated with the electronic messages, adjusting at least one of the associated weight values for at least one of the keywords found in the body of the electronic message and having the predefined association with the determined category having the strongest correlation, wherein the detected user action comprises at least one of replying to the electronic message, archiving the electronic message, and deleting the electronic message.

2. The method of claim 1, wherein the operation further comprises:

displaying the category of the electronic message along with categories of a plurality of other classified electronic messages in an electronic message box;

receiving a request to sort the electronic messages in the electronic message box based on respective categories; and

upon receiving the request, sorting the electronic messages in the electronic message box based on respective categories.

3. The method of claim 1, wherein the operation further comprises:

displaying the relevancy score of the electronic message along with relevancy scores of a plurality of other classified electronic messages in an electronic message box;

receiving a request to sort the electronic messages in the electronic message box based on respective relevancy scores; and

upon receiving the request, sorting the electronic messages in the electronic message box based on respective relevancy scores.

4. The method of claim 1, wherein the at least one of the associated weight values is adjusted to strengthen the correlation between the respective key word and the category upon detecting the user action.

5. The method of claim 1, wherein the at least one of the associated weight values is adjusted to weaken the correlation between the respective key word and the category upon detecting the user action.

6. The method of claim 1, wherein the electronic message is an e-mail.

7. The method of claim 1, wherein the detected user action has a predefined influence on a strength of the at least one of the associated weight values.

8. A computer program product for classifying electronic messages, the computer program product comprising a computer readable storage medium having computer usable program code configured for:

providing access to a predefined list of categories, wherein each of the categories has a predefined association with a respective one or more key words, and wherein each of the one or more key words has an associated weight value reflecting a correlation between the respective key word and the category;

for each category in the predefined list of categories, calculating a relevancy score based on the one or more key words found in a body of the electronic message, wherein the relevancy score reflects a correlation between the electronic message and the category;

associating the electronic message with a category from the predefined list of categories having a strongest correlation with the electronic message based on the calculated relevancy scores; and

responsive to detecting a user action associated with the electronic messages, adjusting at least one of the associated weight values for at least one of the keywords found in the body of the electronic message and having the predefined association with the determined category having the strongest correlation, wherein the detected user action comprises at least one of replying to the electronic message, archiving the electronic message, and deleting the electronic message.

9. The computer program product of claim 8, wherein the operation further comprises:

displaying the category of the electronic message along with categories of a plurality of other classified electronic messages in an electronic message box;

receiving a request to sort the electronic messages in the electronic message box based on respective categories; and

upon receiving the request, sorting the electronic messages in the electronic message box based on respective categories.

10. The computer program product of claim 8, wherein the operation further comprises:

displaying the relevancy score of the electronic message along with relevancy scores of a plurality of other classified electronic messages in an electronic message box;

receiving a request to sort the electronic messages in the electronic message box based on respective relevancy scores; and

upon receiving the request, sorting the electronic messages in the electronic message box based on respective relevancy scores.

11. The computer program product of claim 8, wherein the at least one of the associated weight values is adjusted to strengthen the correlation between the respective key word and the category upon detecting the user action.

12. The computer program product of claim 8, wherein the at least one of the associated weight values is adjusted to weaken the correlation between the respective key word and the category upon detecting the user action.

13. The computer program product of claim 8, wherein the electronic message is an e-mail.

14. The computer program product of claim 8, wherein the detected user action has a predefined influence on a strength of the at least one of the associated weight values.

15. A system, comprising:

a memory comprising an electronic message program; and

at least one processor which, when executing the electronic message program is configured to;

providing access to a predefined list of categories, wherein each of the categories has a predefined association with a respective one or more key words, and wherein each of the one or more key words has an associated weight value reflecting a correlation between the respective key word and the category;

for each category in the predefined list of categories, calculating a relevancy score based on the one or more key words found in a body of the electronic message, wherein the relevancy score reflects a correlation between the electronic message and the category;

associating the electronic message with a category from the predefined list of categories having a strongest correlation with the electronic message based on the calculated relevancy scores; and

responsive to detecting a user action associated with the electronic messages, adjusting at least one of the associated weight values for at least one of the keywords found in the body of the electronic message and having the predefined association with the determined category having the strongest correlation, wherein the detected user action comprises at least one of replying to the electronic message, archiving the electronic message, and deleting the electronic message.

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

display the category of the electronic message along with categories of a plurality of other classified electronic messages in an electronic message box;

receive a request to sort the electronic messages in the electronic message box based on respective categories; and

upon receiving the request, sort the electronic messages in the electronic message box based on respective categories.

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

display the relevancy score of the electronic message along with relevancy scores of a plurality of other classified electronic messages in an electronic message box;

receive a request to sort the electronic messages in the electronic message box based on respective relevancy scores; and

upon receiving the request, sort the electronic messages in the electronic message box based on respective relevancy scores.

18. The system of claim 15, wherein the processor is further configured to adjust the weight value to strengthen the correlation between key word and the category upon detecting the user action.

19. The system of claim 15, wherein the processor is further configured to adjust the weight value to weaken the correlation between key word and the category upon detecting the user action.

20. The system of claim 15, wherein the electronic message is an e-mail.

21. The system of claim 15, wherein the detected user action has a predefined influence on a strength of the at least one of the associated weight values.