Active and passive personalization techniques
View Patent ↗A method for personalizing content for a particular user in a computing system comprising a user interface configured to display content. The method comprises identifying a long term profile having one or more features in a feature set and a long term level of importance associated with each term in the feature set, identifying a short term profile having one or more features in the feature set and a short term level of importance associated with each term in the feature set, identifying input related to the display of the one or more content items on the user interface, and using the input to modify the short term level of importance and the long term level of importance associated with each term in the feature set to form a modified user interest set.
1. In a computer system comprising a user interface configured to display content items, a method for personalizing content for a particular user, the method comprising:
identifying a long term profile having one or more features in a feature set and a long term level of importance associated with each feature in the feature set, the long term level of importance associated with each feature in the feature set being based on a first, longer history of user input received from the user via a social network;
identifying a short term profile having one or more features in the feature set and a short term level of importance associated with each feature in the feature set, the short term level of importance associated with each feature in the feature set being based on a second, shorter history of user input received from the user via the social network;
identifying input related to the display of the one or more content items on the user interface; and
using the input to modify the short term level of importance and the long term level of importance associated with each feature in the feature set to form a modified user interest set.
2. The method as recited in claim 1 , further comprising changing the display of one or more content items based on the modified user interest set.
3. The method as recited in claim 1 , wherein using the input to modify the short term level of importance and the long term level of importance associated with each feature in the feature set comprises modifying the short term level of importance using a different algorithm than used for modifying the long term level of importance.
4. The method as recited in claim 1 , wherein using input to modify the short term level of importance and the long term level of importance associated with each feature in the feature set comprises modifying the short term and long term level of importance based on positive user input using a different algorithm than used for modifying the short term and long term level of importance based on negative user input.
5. The method as recited in claim 1 , wherein the input used to modify the short term level of importance and the long term level of importance is both active input and passive input.
6. The method as recited in claim 5 , further comprising attaching a maximum level of importance to a feature in the feature set which maximum level follows an asymptotic curve for each active or passive inputs that have been received for a particular feature set.
7. The method as recited in claim 1 , wherein the input used to modify the short term level of importance and the long term level of importance modifies the short term level of importance to a greater degree than the long term level of importance.
8. The method as recited in claim 7 , wherein the short term level of importance is modified twice as much as the long term level of importance.
9. The method as recited in claim 1 , wherein at time T 0 , the short term profile and the long term profile are essentially the same.
10. The method as recited in claim 1 , wherein the short and long term profiles include an aging value for determining the level of importance of one or more features of the feature set.
11. The method as recited in claim 1 , wherein the first, longer history of user input received from the user via the social network, upon which the long term level of importance associated with each feature in the feature set being is based, includes a plurality of user selections of approval icons over a first, longer period of time; and
wherein the second, shorter history of user input received from the user via the social network, upon which the short term level of importance associated with each feature in the feature set being is based, includes a plurality of user selections of approval icons over a second, shorter period of time.
12. A computer program product comprising one or more physical computer-readable media having thereon computer-executable instructions that are structured such that, when executed by one or more processors of a computing system including a user interface configured to display content items, the computing system is caused to perform a method a method for personalizing content for a particular user, the method comprising:
identifying a long term profile having one or more features in a feature set and a long term level of importance associated with each feature in the feature set, the long term level of importance associated with each feature in the feature set being based on a first, longer history of user input received from the user via a social network;
identifying a short term profile having one or more features in the feature set and a short term level of importance associated with each feature in the feature set, the short term level of importance associated with each feature in the feature set being based on a second, shorter history of user input received from the user via the social network;
identifying input related to the display of the one or more content items on the user interface; and
using the input to modify the short term level of importance and the long term level of importance associated with each feature in the feature set to form a modified user interest set.
13. The computer program product as recited in claim 12 , wherein the physical computer-readable media has thereon computer-executable instructions that, when executed by the one or more processors, further cause the computing system to perform the following:
changing the display of one or more content items based on the modified user interest set.
14. The computer program product as recited in claim 12 , wherein using the input to modify the short term level of importance and the long term level of importance associated with each feature in the feature set comprises modifying the short term level of importance using a different algorithm than used for modifying the long term level of importance.
15. The computer program product as recited in claim 12 , wherein using input to modify the short term level of importance and the long term level of importance associated with each feature in the feature set comprises modifying the short term and long term level of importance based on positive user input using a different algorithm than used for modifying the short term and long term level of importance based on negative user input.
16. The computer program product as recited in claim 12 , wherein the input used to modify the short term level of importance and the long term level of importance is both active input and passive input.
17. The computer program product as recited in claim 12 , wherein the input used to modify the short term level of importance and the long term level of importance modifies the short term level of importance to a greater degree than the long term level of importance.
18. The computer program product as recited in claim 17 , wherein the short term level of importance is modified twice as much as the long term level of importance.
19. The computer program product as recited in claim 12 , wherein at time T 0 , the short term profile and the long term profile are essentially the same.
20. The computer program product as recited in claim 12 , wherein the short and long term profiles include an aging value for determining the level of importance of one or more features of the feature set.
21. The computer program product as recited in claim 12 , wherein the physical computer-readable media includes one of RAM, ROM, EEPROM, CD-ROM, other optical disk storage, magnetic disk storage or other magnetic storage devices.
22. The computer program product as recited in claim 12 , wherein the first, longer history of user input received from the user via the social network, upon which the long term level of importance associated with each feature in the feature set being is based, includes a plurality of user selections of approval icons over a first, longer period of time; and
wherein the second, shorter history of user input received from the user via the social network, upon which the short term level of importance associated with each feature in the feature set being is based, includes a plurality of user selections of approval icons over a second, shorter period of time.