Automatic adjustment of email filters based on browser history and telecommunication records
View Patent ↗The present invention, in one embodiment, is directed to a spam filter 132 that identifies, based on a likely interest of a user, electronic text messages addressed to the user as being spam or non-spam and determines a likely interest of the user using at least one of the following: (i) at least part of a search query provided to a search engine; (ii) content of at least one web page selected by the user during a browsing session; (iii) content of a cookie; (iv) a live voice electronic address; and (v) an entry in an electronic calendar of the user.
1. A method, comprising:
(a) receiving an electronic text message addressed to a user;
(b) determining a likely interest of the user in the electronic text message using at least one of the following objects:
at least part of a search query provided to a search engine;
content of a cookie;
a live voice electronic address; and
an entry in an electronic calendar of the user;
wherein a persistency life is assigned to the at least one of the objects, wherein the persistency life indicates a temporal life of the at least one of the following objects, and wherein the at least one of the following objects is deleted when the temporal life has expired; and
(c) based on a determined likely interest of the user, determining whether the electronic text message is spam or non-spam.
2. The method of claim 1 , wherein at least part of a search query provided to a search engine is used to determine a likely interest of the user.
3. The method of claim 2 , wherein a probative value of the search query to an interest of the user is based on at least one of a duration of the search, a frequency of use by the user of the search terms, a web site selected by the user during the search, and an outcome of the search.
4. The method of claim 1 , wherein a content of at least one web page selected by the user during a browsing session in addition to the at least one of the objects is used to determine a likely interest of the user.
5. The method of claim 4 , wherein the content is a tag associated with a link selected by the user.
6. The method of claim 1 , wherein a content of a cookie is used to determine a likely interest of the user.
7. The method of claim 1 , wherein a live voice electronic address is used to determine a likely interest of the user and wherein the electronic address is a telephone number.
8. The method of claim 1 , wherein an entry in an electronic calendar is used to determine a likely interest of the user.
9. The method of claim 1 , wherein, in step (b), a probative value is assigned to the at least one of the objects, wherein a probative value is used to adjust a HAM score assigned to the message, the HAM score indicating a likelihood that the message is spam or non-spam.
10. A non-transitory computer readable medium comprising processor-executable instructions for performing the steps of claim 1 .
11. A system, comprising:
a spam filter operating on a processor to identify based on a likely interest of a user, electronic text messages addressed to the user as being spam or non-spam and determine a likely interest of the user using at least one of the following objects:
at least part of a search query provided to a search engine;
content of a cookie;
a live voice electronic address; and
an entry in an electronic calendar of the user,
wherein a persistency life is assigned to the at least one of the following objects that was used to determine the likely interest of the user, wherein the persistency life is a measure of a time period over which the at least one of the following objects is deemed to be probative of user interest and wherein the at least one of the following objects are removed from consideration when the persistency life expires.
12. The system of claim 11 , wherein at least part of a search query is used to determine a likely interest of the user.
13. The system of claim 12 , wherein a probative value of the search query to an interest of the user is based on at least one of a duration of the search, a frequency of use by the user of the search terms, a web site selected by the user during the search, and an outcome of the search.
14. The system of claim 11 , wherein in addition to the at least one of the objects a content of at least one web page selected by the user during a browsing session is used to determine a likely interest of the user.
15. The system of claim 14 , wherein the content is a tag associated with a link selected by the user.
16. The system of claim 11 , wherein a content of a cookie is used to determine a likely interest of the user.
17. The system of claim 11 , wherein a live voice electronic address is used to determine a likely interest of the user and wherein the electronic address is a telephone number.
18. The system of claim 11 , wherein an entry in an electronic calendar of the user is used to determine a likely interest of the user.
19. The system of claim 11 , wherein a probative value is used to adjust a HAM score assigned to the message, the HAM score indicating a likelihood that the message is spam or non-spam.