System for reclassification of electronic messages in a spam filtering system

• US 9,961,029 B2
• Filed: 07/09/2014
• Issued: 05/01/2018
• Est. Priority Date: 06/30/2004
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google LLC (Alphabet Inc.)
• Inventors: Dulitz, Daniel Wesley, Golub, Seth J.
• Primary Examiner(s): Chen, Eric
• Application Number:
  US14/327,478
• Publication Number:
  US 20140325007A1
• Time in Patent Office: 1,392 Days
• 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

Abstract

Systems and methods are provided that, in response to obtaining an email to a recipient from a sender, and in accordance with a determination that an indirect relationship exists between the sender and the recipient, determine a spam probability of the email by evaluating statistical information regarding the historical electronic interactions associated with the sender. In this way, the email is classified according to the identified spam probability.

Claims

1. A method comprising:

obtaining an email that is sent from a sender to a recipient; and

in response to determining that no direct relationship exists between the sender and the recipient, and instead that an indirect relationship exists between the sender and the recipient;

determining, based at least in part on the indirect relationship, a spam probability associated with the email by evaluating statistical information regarding historical electronic interactions associated with the sender and the recipient, wherein the historical electronic interactions associated with the sender and the recipient includes historical electronic interactions from one or more non-email applications, and wherein the statistical information includes information associated with a frequency of interaction between the sender and the recipient; and

classifying the email according to the identified spam probability,

wherein the direct relationship is defined by a historical bi-directional interaction between the sender and the recipient in which the recipient has previously accepted and replied to at least one electronic message sent by the sender, and

wherein the indirect relationship is defined only by a historical unidirectional interaction between the sender and the recipient in which the recipient has not previously accepted and replied to any email messages sent by the sender.

2. The method of claim 1, wherein determining the spam probability further comprises:

combining historical electronic interactions from an email application and the historical electronic interactions from one or more non-email applications to produce a combined condition; and

identifying the spam probability of the email based at least in part on the combined condition.

3. The method of claim 2, wherein the one or more non-email applications includes one or more of an instant messaging application, an on-line chat application, a file-sharing application, or a collaboration application.

4. The method of claim 1, wherein identifying the spam probability further comprises:

identifying the spam probability of the email based on historical electronic interactions between one or more additional recipients and the sender.

5. The method of claim 1, wherein:

the statistical information regarding historical interactions associated with the sender includes information associated with types of interaction.

6. A system comprising:

at least one processor; and

a memory storing one or more programs to be executed by the at least one processor, the one or more programs comprising instructions to;

obtain an email that is sent from a sender to a recipient; and

in response to determining that no direct relationship exists between the sender and the recipient, and instead that an indirect relationship exists between the sender and the recipient;

determine, based at least in part on the indirect relationship, a spam probability associated with the email by evaluating statistical information regarding historical electronic interactions associated with the sender and the recipient, wherein the historical electronic interactions associated with the sender and the recipient includes historical electronic interactions from one or more non-email applications, and wherein the statistical information includes information associated with a frequency of interaction between the sender and the recipient; and

classify the email according to the identified spam probability,

wherein the direct relationship is defined by a historical bi-directional interaction between the sender and the recipient in which the recipient has previously accepted and replied to at least one electronic message sent by the sender, and

wherein the indirect relationship is defined only by a historical unidirectional interaction between the sender and the recipient in which the recipient has not previously accepted and replied to any email messages sent by the sender.

7. The system of claim 6, wherein the instructions to determine the spam probability further comprise instructions to:

combine historical electronic interactions from an email application and the historical electronic interactions from one or more non-email applications to produce a combined condition; and

identify the spam probability of the email based at least in part on the combined condition.

8. The system of claim 7, wherein the one or more non-email applications also includes one or more of an instant messaging application, an on-line chat application, a file-sharing application, or a collaboration application.

9. The system of claim 6, wherein the instructions to identify the spam probability further comprise instructions to:

identify the spam probability of the email based on historical electronic interactions between one or more additional recipients and the sender.

10. The system of claim 6, wherein:

the statistical information regarding historical interactions associated with the sender includes one or more of information associated with types of interaction or information associated with a media of interaction.

11. A non-transitory computer readable storage medium storing one or more programs configured for execution by a computer, the one or more programs comprising instructions to:

obtain an email that is sent from a sender to a recipient; and

in response to determining no direct relationship exists between the sender and the recipient, and instead that an indirect relationship exists between the sender and the recipient;

determine, based at least in part on the indirect relationship, a spam probability associated with the email by evaluating statistical information regarding historical electronic interactions associated with the sender and the recipient, wherein the historical electronic interactions associated with the sender and the recipient includes historical electronic interactions from one or more non-email applications, and wherein the statistical information includes information associated with a frequency of interaction between the sender and the recipient; and

classify the email according to the identified spam probability,

wherein the direct relationship is defined by a historical bi-directional interaction between the sender and the recipient in which the recipient has previously accepted and replied to at least one electronic message sent by the sender, and

wherein the indirect relationship is defined only by a historical unidirectional interaction between the sender and the recipient in which the recipient has not previously accepted and replied to any email messages sent by the sender.

12. The non-transitory computer readable storage medium of claim 11, wherein the instructions to determine the spam probability further comprise instructions to:

combine historical electronic interactions from an email application and the historical electronic interactions from one or more non-email applications to produce a combined condition; and

identify the spam probability of the email based at least in part on the combined condition.

13. The non-transitory computer readable storage medium of claim 12, wherein the one or more non-email applications includes one or more of an instant messaging application, an on-line chat application, a file-sharing application, or a collaboration application.

14. The non-transitory computer readable storage medium of claim 11, wherein the instructions to identify the spam probability further comprise instructions to:

identifying the spam probability of the email based on historical electronic interactions between one or more additional recipients and the sender.

15. The non-transitory computer readable storage medium of claim 11, wherein:

the statistical information regarding historical interactions associated with the sender includes information associated with types of interaction.