Method and apparatus for providing feedback for email filtering

• US 20050065906A1
• Filed: 08/11/2004
• Published: 03/24/2005
• Est. Priority Date: 08/19/2003
• Status: Abandoned Application

• Current Assignee: Wizaz KK
• Original Assignee: Wizaz KK
• Inventors: Romero, Timothy L.
• Application Number:
  US10/915,690
• Publication Number:
  US 20050065906A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06Q 10/107 Computer-aided management of electronic mailing [e-mailing]
• CPC Class Codes: H04L 51/214 using selective forwarding

Abstract

The invention provides a novel system for email users to provide feedback to email routing, filtering and classification systems. The invention uses email generated by standard email client software as the transport mechanism for providing this feedback, and thereby eliminates the need for custom, client-side software to be installed on the user'"'"'s computer.

Claims

1. A method comprising:

providing feedback to a classifier

creating a database using a first electronic communication processed by the classifier,

forwarding a second electronic communication based on the first electronic communication to a specified mailbox to be used as a feedback example,

extracting header information from the second electronic communication,

using the extracted header information to retrieve the first electronic communication from the database, and

training the classifier using the first electronic communication as an example of a category indicated by the specified mailbox at which the second electronic communication was received.

2. The method of claim 1, in which the creating of a database comprises deriving statistics from the first electronic communication and storing the derived statistics in the database.

3. The method of claim 1, further comprising attaching the first electronic communication to the second electronic communication.

4. The method of claim 3, in which the attached first electronic communication is re-analyzed by the classifier.

5. The method of claim 1, further comprising indicating the category with a text command that appears at a predefined location in the second electronic communication.

6. The method of claim 1, further comprising creating the second electronic communication using a dedicated user interface.

7. The method of claim 3, further comprising creating the second electronic communication using a dedicated user interface.

8. The method of claim 5, in which the predefined location is in a body of the second electronic communication.

9. The method of claim 5, further comprising sending the second electronic communication to the same mailbox that is checked by the classifier, and only an electronic communication having said command is processed as a second electronic communication.

10. The method of claim 5, further comprising indicating the category by providing the word “

category”

or one of its synonyms on a first line of the second electronic communication.

11. The method of claim 10, further comprising indicating the category by providing information in a subject line of the second electronic communication.

12. The method of claim 10, further comprising indicating the category by providing information in the header information of the second electronic communication.

13. The method of claim 5, further comprising providing additional security by providing a password in a body of the second electronic communication.

14. The method of claim 1, in which the first electronic communication and the second electronic communication are email.

15. A method for storing and retrieving an electronic communication or information derived from the electronic communication, the method comprising:

storing information derived from an electronic communication by,

creating an index based on header information of the electronic communication,

removing non-essential and descriptive information from the header information,

storing the remaining information such that it is linked to said index, and

retrieving the stored information by,

forwarding the electronic communication to a designated mailbox,

extracting the original electronic communication'"'"'s header information from a header block of the forwarded electronic communication, and

retrieving the information based on this these extracted headers.

16. The method of claim 15, further comprising storing a complete copy of the electronic communication.

17. The method of claim 15, further comprising storing statistical information derived from the original electronic communication.

18. The method of claim 15, further comprising sending information derived from the original electronic communication as an attachment to the forwarded electronic communication, and extracting the header information from the header of the attached electronic communication rather than the header block of the forwarded electronic communication.

19. The method of claim 15, further comprising converting a Date header stored in the index and a date information extracted from the header block of the forwarded electronic communication to a common time zone.

20. The method of claim 15, further comprising extracting either a Sent or a Date information from the header block and matching the extracted information to respective indexed Sent or Date fields in the header.

21. The method of claim 15, in which a date and time of the forwarded electronic communication will be considered a match to an indexed Date field if the forwarded electronic communication contains seconds information and it matches to the second, the date and time of the forwarded electronic communication will be considered a match to the index Date field if the forwarded electronic communication does not contain seconds and it matches to the minute.

22. The method of claim 15, further comprising setting the extracted date'"'"'s time zone to a time zone of the training electronic communication if the time zone information is missing.

23. The method of claim 15, further comprising storing the non-essential information in the index and using the non-essential information to retrieve the stored information.

24. The method of claim 15, in which an extracted From field is considered to match an index entry if it matches a From or Sender field of the electronic communication.

25. The method of claim 15, wherein if either a To or a From header cannot be extracted from the header block, then the field that is extracted is used in conjunction with a Date field to match the index.

26. The method of claim 15, wherein the header information of the electronic communication comprises Date, To, From, and Sender information.

27. A system to provide feedback to an classifier, the system comprising:

a classifier to classify received electronic communications,

a database to store received electronic communication information, and

a plurality of user mailboxes to allow users to access electronic communications,

wherein

the classifier receives a first electronic communication,

the classifier stores information relating to the first electronic communication in the database,

the classifier constructs an index of the stored information based on a header of the first electronic communication,

the classifier forwards the first electronic communication to one of the plurality of user mailboxes,

a user determines if the first electronic communication is to be used to train the classifier,

if the first electronic communication is to be used to train the classifier, the user provides a second electronic communication containing information about the first electronic communication to the classifier, and

the classifier updates one of a plurality of classification filters based on the second electronic communication.

28. The system according to claim 27 wherein the electronic communications are email.

29. The system according to claim 27 wherein the information relating to the first electronic communication that is stored in the database comprises the complete text of the first electronic communication.

30. The system according to claim 27 wherein the information relating to the first electronic communication that is stored in the database comprises statistical information derived from the first electronic communication.

31. The system according to claim 27 wherein the information relating to the first electronic communication that is stored in the database comprises the body of the first electronic communication.

32. The system according to claim 27 wherein the information relating to the first electronic communication that is stored in the database comprises header information.

33. The system according to claim 27 further comprising providing the second electronic communication to the classifier by sending the second electronic communication to a general control mailbox.

34. The system according to claim 27 further comprising providing the second electronic communication to the classifier by sending the training electronic communication to a specific control mailbox.

35. The system according to claim 27 further comprising attaching a copy of the first electronic communication to the second electronic communication.

36. The system according to claim 27, wherein the classifier retrieves the first electronic communication in response to the second electronic communication.

37. The system according to claim 27 wherein the classifier analyses the first electronic communication in response to the second electronic communication.

38. The system according to claim 33 further comprising determining which of the plurality of classification filters is to be updated according to a text command located at a predetermined location in the second electronic communication.

39. The system according to claim 38 wherein the general control mailbox processes electronic communications containing the text command as second electronic communications, and processes electronic communications not containing the text command as first electronic communications.

40. The system according to claim 38 wherein the text command is located in the body of the second electronic communication.

41. The system according to claim 38 wherein the text command includes the word category, and is located on the first line of the second electronic communication.

42. The system according to claim 38 wherein the text command is located in a subject line of the second electronic communication.

43. The system according to claim 38 wherein the text command is located in the header of the second electronic communication.

44. The system according to claim 34 further comprising updating the one of the plurality of classification filters according to the specific control mailbox to which the second electronic communication is sent.

45. The system according to claim 27 wherein the second electronic communication is generated by a dedicated user interface.

46. The system according to claim 27 wherein the classifier updates one of a plurality of classification filters only if the second electronic communication is an authorized second electronic communication.

47. The system according to claim 46 wherein authorized second electronic communications are authenticated using a password.

48. A method for training a classifier comprising:

receiving a first electronic communication,

storing information associated with the first electronic communication,

forwarding the first electronic communication to a user,

receiving a second electronic communication from a user, and

updating one of a plurality of classification filters based on the second electronic communication.

49. The method of claim 48 wherein the information regarding the first electronic communication is stored in a database.

50. The method of claim 48 wherein the stored information is a body of the first electronic communication.

51. The method of claim 49, wherein the stored information is statistical information derived from the first electronic communication.

52. The method of claim 48, further comprising attaching the first electronic communication to the second electronic communication.

53. The method of claim 49, further comprising including header information from the first electronic communication in the second electronic communication.

54. The method of claim 53, further comprising retrieving the stored information from the database based on the header information included in the second electronic communication.

55. The method of claim 54, wherein the updating one of a plurality of classification filters based on the second electronic communication is performed using the retrieved information.

56. The method of claim 49, further comprising creating an index of the information stored in the data base.

57. The method of claim 56, wherein the index is created using header information.

58. The method of claim 48, wherein the updating one of a plurality of classification filters based on the second electronic communication comprises:

reducing the header information of the first electronic communication into a generic format regardless of which of a plurality of electronic communication clients has been used, and using the reduce header information to update the classification filter.

59. The method of claim 48, wherein the second electronic communication and the first electronic communication are both received by a general mailbox.

60. The method of claim 48, wherein the second electronic communication is received by a specific mailbox related to a particular one of the plurality of classification filters.

61. The method of claim 48, wherein the electronic communications are emails.

62. A method of training an electronic communication classifier comprising:

receiving a first electronic communication, the first electronic communication comprising a plurality of example communications,

extracting the plurality of example communications from the first electronic communication, and

modifying one of a plurality of classification filters based on the extracted example communications.

63. The method of claim 62, wherein the first electronic communication and the example communications are email.

64. The method of claim 62, wherein the first electronic communication is received by a general control mailbox.

65. The method of claim 62, wherein the first electronic communication is received by a specific control mailbox.

66. The system according to claim 62, wherein the classifier updates the one of a plurality of classification filters only if the first electronic communication is authorized.

67. The system according to claim 62, further comprising authorizing second electronic communications using a password.