Electronic mail determination method and system and storage medium

• US 6,199,103 B1
• Filed: 06/23/1998
• Issued: 03/06/2001
• Est. Priority Date: 06/24/1997
• Status: Expired due to Term

• Current Assignee: Tumbleweed Holdings LLC (Equitable IP Corporation)
• Original Assignee: Omron Corporation
• Inventors: Sawada, Akira, Fujii, Fujiki, Sakaguchi, Manabu, Hori, Masaki
• Primary Examiner(s): Maung, Zarni
• Assistant Examiner(s): LEE, TAK KAM
• Application Number:
  US09/102,768
• Time in Patent Office: 987 Days
• Field of Search: 706/47, 709/206, 709/207, 709/205, 709/203, 370/401, 370/465
• US Class Current: 709/206
• 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

The invention is directed to a system that can easily generate determination conditions and determine a plurality of similar junk electronic mail pieces based on one determination condition. An electronic mail determination system comprises a junk electronic mail determination processing section (2) for determining whether or not a given electronic mail piece through an input section (1) is junk based on the determination condition stored in a junk electronic mail determination condition storage section (3), an estimated junk electronic mail storage section (6) for storing the electronic mail piece if the electronic mail piece is determined junk by the junk electronic mail determination processing section, and a junk electronic mail exemplification learning section (7) for analyzing content information of the electronic mail piece stored in the storage section (6) for extracting a feature amount to determine that electronic mail is junk and adding the extracted feature amount to the junk electronic mail determination condition storage section as a junk electronic mail determination condition. A keyword vector is used as the determination condition.

Claims

1. An electronic mail determination method comprising the steps of:

determining whether or not a given electronic mail piece is junk based on a determination condition stored in a junk electronic mail determination condition storage section;

if the electronic mail piece is determined junk, storing it in an estimated junk electronic mail storage section;

reading therein electronic mail piece stored in the estimated junk electronic mail storage section and analyzing contents of entire text of said electronic mail piece for extracting a feature amount to determine that said electronic mail piece is junk; and

adding the extracted feature amount to the junk electronic mail determination condition storage section as a junk electronic mail determination condition.

2. The electronic mail determination method as claimed in claim 1 wherein said determining step determines whether or not the determination condition is similar to the feature amount extracted from the given electronic mail piece.

3. The electronic mail determination method as claimed in claim 1 or 2 further comprising the steps of:

storing an electronic mail piece determined non-junk at said determining step in an estimated non-junk electronic mail storage section;

reading the electronic mail piece stored in the estimated non-junk electronic mail storage section and analyzing content information of the electronic mail piece for extracting a feature amount to determine that electronic mail is non-junk; and

adding the extracted feature amount to the non-junk electronic mail determination condition storage section as a non-junk electronic mail determination condition, wherein

said determining step determines whether or not the given electronic mail piece is junk based on the determination condition stored in the junk electronic mail determination condition storage section and the determination condition stored in the non-junk electronic mail determination condition storage section.

4. The electronic mail determination method as claimed in any one of the preceding claims further comprising the step of automatically deleting electronic mail stored in the estimated junk electronic mail storage section after the expiration of a predetermined time interval from the determining step time.

5. The electronic mail determination method as claimed in claim 3 further comprising the steps of:

extracting a feature amount of the electronic mail piece stored in the estimated non-junk electronic mail storage section;

storing the extracted feature amount in a non-junk electronic mail feature amount storage section; and

subtracting the feature amount stored in the non-junk electronic mail feature amount storage section from feature amount generated based on junk electronic mail stored in the junk electronic mail storage section and storing the resultant feature amount in the junk electronic mail determination condition storage section as a junk electronic mail determination condition.

16. The electronic mail determination method as claimed in claim 4 further comprising the steps of:

extracting a feature amount of the electronic mail piece stored in the estimated non-junk electronic mail storage section;

storing the extracted feature amount in a non-junk electronic mail feature amount storage section; and

subtracting the feature amount stored in the non-junk electronic mail feature amount storage section from feature amount generated based on junk electronic mail stored in the junk electronic mail storage section and storing the resultant feature amount in the junk electronic mail determination condition storage section as a junk electronic mail determination condition.

6. An electronic mail determination system comprising:

a junk electronic mail determination condition storage section for storing a determination condition to determine whether or not electronic mail is junk;

a junk electronic mail determination processing section for determining whether or not a given electronic mail piece is junk based on the determination condition stored in said junk electronic mail determination condition storage section;

an estimated junk electronic mail storage section for storing the electronic mail piece if the electronic mail piece is determined junk by said junk electronic mail determination processing section; and

a junk electronic mail exemplification learning section for analyzing contents of entire text of said electronic mail piece stored in said estimated junk electronic mail storage section for extracting a feature amount to determine that said electronic mail piece is junk and adding the extracted feature amount to said junk electronic mail determination condition storage section as a junk electronic mail determination condition.

7. The electronic mail determination system as claimed in claim 6 wherein said junk electronic mail exemplification learning section generates a determination condition to determine a similar junk electronic mail piece, and wherein

said junk electronic mail determination processing section determines whether or not the determination condition is similar to the feature amount extracted from the given electronic mail piece.

8. The electronic mail determination system as claimed in claim 6 or 7 further including:

an estimated non-junk electronic mail storage section for storing the electronic mail piece determined non-junk by said junk electronic mail determination processing section; and

a non-junk electronic mail exemplification learning section for analyzing content information of the electronic mail piece stored in said estimated non-junk electronic mail storage section for extracting a feature amount to determine that electronic mail is non-junk and adding the extracted feature amount to said non-junk electronic mail determination condition storage section as a non-junk electronic mail determination condition, wherein

said junk electronic mail determination processing section determines whether or not the given electronic mail piece is junk based on the determination condition stored in said junk electronic mail determination condition storage section and the determination condition stored in said non-junk electronic mail determination condition storage section.

9. The electronic mail determination system as claimed in claim 6 or 7 further including a deletion processing section for automatically deleting electronic mail stored in said estimated junk electronic mail storage section after the expiration of a predetermined time interval from the determination processing time.

10. The electronic mail determination system as claimed in claim 8 further including:

a non-junk electronic mail feature amount extraction section for extracting a feature amount of the electronic mail piece stored in said estimated non-junk electronic mail storage section; and

a non-junk electronic mail feature amount storage section for storing the extracted feature amount, wherein

said junk electronic mail exemplification learning section subtracts the feature amount stored in said non-junk electronic mail feature amount storage section from feature amount generated based on junk electronic mail stored in said junk electronic mail storage section and adopts the resultant feature amount as a junk electronic mail determination condition.

17. The electronic mail determination system as claimed in claim 8 further including a deletion processing section for automatically deleting electronic mail stored in said estimated junk electronic mail storage section after the expiration of a predetermined time internal from the determination processing time.

18. The electronic mail determination system as claimed in claim 9 further including:

a non-junk electronic mail feature amount extraction section for extracting a feature amount of the electronic mail piece stored in said estimated non-junk electronic mail storage section; and

a non-junk electronic mail feature amount storage section for storing the extracted feature amount, wherein

said junk electronic mail exemplification learning section subtracts the feature amount stored in said non-junk electronic mail feature amount storage section from feature amount generated based on junk electronic mail stored in said junk electronic mail storage section and adopts the resultant feature amount as a junk electronic mail determination condition.

11. A computer-readable recording medium storing an electronic mail determination program containing instructions for causing a computer to execute the steps of:

determining whether or not a given electronic mail piece is junk based on a determination condition stored in a junk electronic mail determination condition storage section;

if the electronic mail piece is determined junk, storing it in an estimated junk electronic mail storage section;

reading the electronic mail piece stored in the estimated junk electronic mail storage section and analyzing contents of entire text of said electronic mail piece for extracting a feature amount to determine that said electronic mail piece is junk; and

adding the extracted feature amount to the junk electronic mail determination condition storage section as a junk electronic mail determination condition.

12. The computer-readable recording medium as claimed in claim 11 wherein said determining step is a program for determining whether or not the determination condition is similar to the feature amount extracted from the given electronic mail piece.

13. The computer-readable recording medium as claimed in claim 11 or 12 further storing a program for causing a computer to execute the steps of:

storing an electronic mail piece determined non-junk at said determining step in an estimated non-junk electronic mail storage section;

reading the electronic mail piece stored in the estimated non-junk electronic mail storage section and analyzing content information of the electronic mail piece for extracting a feature amount to determine that electronic mail is non-junk; and

adding the extracted feature amount to the non-junk electronic mail determination condition storage section as a non-junk electronic mail determination condition, wherein

said determining step determines whether or not the given electronic mail piece is junk based on the determination condition stored in the junk electronic mail determination condition storage section and the determination condition stored in the non-junk electronic mail determination condition storage section.

14. The computer-readable recording medium as claimed in claim 11 or 12 further storing a program for causing a computer to execute the step of automatically deleting electronic mail stored in the estimated junk electronic mail storage section after the expiration of a predetermined time interval from the determining step time.

15. The computer-readable recording medium as claimed in claim 13 further storing a program for causing a computer to execute the steps of:

extracting a feature amount of the electronic mail piece stored in the estimated non-junk electronic mail storage section;

storing the extracted feature amount in a non-junk electronic mail feature amount storage section; and

subtracting the feature amount stored in the non-junk electronic mail feature amount storage section from feature amount generated based on junk electronic mail stored in the junk electronic mail storage section and storing the resultant feature amount in the junk electronic mail determination condition storage section as a junk electronic mail determination condition.

19. The computer-readable recording medium as claimed in claim 13 further storing a program for causing a computer to execute the step of automatically deleting electronic mail stored in the estimated junk electronic mail storage section after the expiration of a predetermined time interval from the determining step time.

20. The computer-readable recording medium as claimed in claim 14 further storing a program for causing a computer to execute the steps of:

extracting a feature amount of the electronic mail piece stored in the estimated non-junk electronic mail storage section;

storing the extracted feature amount in a non-junk electronic mail feature amount storage section; and

subtracting the feature amount stored in the non-junk electronic mail feature amount storage section from feature amount generated based on junk electronic mail stored in the junk electronic mail storage section and storing the resultant feature amount in the junk electronic mail determination condition storage section as a junk electronic mail determination condition.