ONLINE ADAPTIVE FILTERING OF MESSAGES

• US 20130007152A1
• Filed: 07/03/2012
• Published: 01/03/2013
• Est. Priority Date: 07/21/2003
• Status: Active Grant

• Current Assignee: Yahoo Assets LLC
• Original Assignee: AOL Inc. (Apollo Global Management, Inc.)
• Inventors: Alspector, Joshua, Kolcz, Aleksander
• Granted Patent: US 8,799,387 B2
• Time in Patent Office: Days
• US Class Current: 709/206
• CPC Class Codes: G06F 16/353 into predefined classes
• CPC Class Codes: G06Q 10/107 Computer-aided management of electronic mailing [e-mailing]
• CPC Class Codes: H04L 51/212 using filtering or selective blocking
• CPC Class Codes: H04L 51/48 Message addressing, e.g. address format or anonymous messages, aliases

Abstract

In general, a two or more stage spam filtering system is used to filter spam in an e-mail system. One stage includes a global e-mail classifier that classifies e-mail as it enters the e-mail system. The parameters of the global e-mail classifier generally may be determined by the policies of e-mail system owner and generally are set to only classify as spam those e-mails that are likely to be considered spam by a significant number of users of the e-mail system. Another stage includes personal e-mail classifiers at the individual mailboxes of the e-mail system users. The parameters of the personal e-mail classifiers generally are set by the users through retraining, such that the personal e-mail classifiers are refined to track the subjective perceptions of their respective user as to what e-mails are spam e-mails.

Claims

1-15. -15. (canceled)

16. A method of operating a spam filtering system in a messaging system that includes a message gateway and individual message boxes for users of the system, wherein a global, scoring e-mail classifier classifies messages coming into the message gateway as spam or non-spam and personal, scoring e-mail classifiers classify messages delivered to the individual message boxes after passing through the global, scoring e-mail classifier, the method comprising:

aggregating personal retraining data used to retrain the personal, scoring e-mail classifiers, wherein personal retraining data for an individual message box is based on a user'"'"'s feedback about the classes of messages in the user'"'"'s individual message box;

selecting a subset of the aggregated personal retraining data as global retraining data; and

retraining the global, scoring e-mail classifier based on the global retraining data so as to adjust which messages are classified as spam.

17. The method of claim 16 wherein the user feedback is explicit.

18. The method of claim 17 wherein the explicit user feedback comprises one or more of the following:

a user reporting a message as spam;

moving a message from an Inbox folder in the individual message box to a Spam folder in the individual message box;

or moving a message from an Spam folder in the individual message box to a Inbox folder in the individual message box.

19. The method of claim 16 wherein the feedback is implicit.

20. The method of claim 19 wherein the implicit feedback comprises one or more of the following:

keeping a message as new after the message has been read;

forwarding a message;

replying to a message;

printing a message;

adding a sender of a message to an address book;

or not explicitly changing a classification of a message.

21. The method of claim 16 wherein the aggregated personal retraining data comprises messages.

22. The method of claim 21 wherein selecting a subset of the aggregated personal retraining data comprises:

determining a difference between a probability measure calculated for a message by the global, scoring e-mail classifier and a classification threshold of the global, scoring e-mail classifier; and

selecting the message as global retraining data if a magnitude of the difference exceeds a threshold difference.

23. The method of claim 16 wherein the feedback comprises changing a message'"'"'s class.

24. The method of claim 23 wherein selecting a subset of the aggregated personal retraining data comprises selecting a message as global retraining data when a particular number of users change the message'"'"'s classification.

25. The method of claim 16 wherein the messages are e-mails.

26. The method of claim 16 wherein the messages are e-instant messages.

27. The method of claim 16 wherein the messages are SMS messages.

28. The method of claim 16 wherein, to classify a message, the global, scoring e-mail classifier uses an internal model to determine a probability measure for the message and compares the probability measure to a classification threshold.

29. The method of claim 28 wherein, to classify a message, the personal, scoring e-mail classifier uses an internal model to determine a probability measure for the message and compares the probability measure to a classification threshold, the method further comprising initializing the personal, scoring e-mail classifier'"'"'s internal model using the internal model for the global, scoring e-mail classifier.

30-43. -43. (canceled)

44. A computer-usable medium storing a computer program for operating a spam filtering system in a messaging system that includes a message gateway and individual message boxes for users of the system, wherein a global, scoring e-mail classifier classifies messages coming into the message gateway as spam or non-spam and personal, scoring e-mail classifiers classify messages delivered to the individual message boxes after passing through the global, scoring e-mail classifier, the computer program comprising instructions for causing a computer to:

aggregate personal retraining data used to retrain the personal, scoring e-mail classifiers, wherein personal retraining data for an individual message box is based on a user'"'"'s feedback about the classes of messages in the user'"'"'s individual message box;

select a subset of the aggregated personal retraining data as global retraining data; and

retrain the global, scoring e-mail classifier based on the global retraining data so as to adjust which messages are classified as spam.

45. The medium of claim 44 wherein the user feedback is explicit.

46. The medium of claim 45 wherein the explicit user feedback comprises one or more of the following:

a user reporting a message as spam;

moving a message from an Inbox folder in the individual message box to a Spam folder in the individual message box;

or moving a message from an Spam folder in the individual message box to a Inbox folder in the individual message box.

47. The medium of claim 44 wherein the feedback is implicit.

48. The medium of claim 47 wherein the implicit feedback comprises one or more of the following:

keeping a message as new after the message has been read;

forwarding a message;

replying to a message;

printing a message;

adding a sender of a message to an address book;

or not explicitly changing a classification of a message.

49. The medium of claim 44 wherein the aggregated personal retraining data comprises messages.

50. The medium of claim 49 wherein, to select a subset of the aggregated personal retraining data, the computer program further comprises instructions for causing a processor to:

determine a difference between a probability measure calculated for a message by the global, scoring e-mail classifier and a classification threshold of the global, scoring e-mail classifier; and

select the message as global retraining data if a magnitude of the difference exceeds a threshold difference.

51. The medium of claim 44 wherein the feedback comprises changing a message'"'"'s class.

52. The medium of claim 51 wherein to select a subset of the aggregated personal retraining data, the computer program further comprises instructions for causing a processor to select a message as global retraining data when a particular number of users change the message'"'"'s classification.

53. The medium of claim 44 wherein the messages are e-mails.

54. The medium of claim 44 wherein the messages are e-instant messages.

55. The medium of claim 44 wherein the messages are SMS messages.

56. The medium of claim 44 wherein, to classify a message, the global, scoring e-mail classifier uses an internal model to determine a probability measure for the message and compares the probability measure to a classification threshold.

57. (canceled)

58. An apparatus for operating a spam filtering system in a messaging system that includes a message gateway and individual message boxes for users of the system, wherein a global, scoring e-mail classifier classifies messages coming into the message gateway as spam or non-spam and personal, scoring e-mail classifiers classify messages delivered to the individual message boxes after passing through the global, scoring e-mail classifier, the apparatus comprising:

means for aggregating personal retraining data used to retrain the personal, scoring e-mail classifiers, wherein personal retraining data for an individual message box is based on a user'"'"'s feedback about the classes of messages in the user'"'"'s individual message box;

means for selecting a subset of the aggregated personal retraining data as global retraining data; and

means for retraining the global, scoring e-mail classifier based on the global retraining data so as to adjust which messages are classified as spam.