System and method for selecting information for display based on past user interactions
A system and method is provided that automatically predicts the link that a user would like to follow by collecting and modeling user actions, and navigates to the link. In one aspect, the system may store a history of the user's manipulation of a browser's user interface, and predict the link of greatest interest to the user by comparing the visual similarity of potential links with previously selected and non-selected links, and estimating the location of the screen that the user is currently viewing.
1. A computer-implemented method of selecting links, comprising:
providing for display to a user a first document at a first location on a network, the first document containing a plurality of links to other documents at other locations of the network where each link in the plurality of links is (1) displayed in a document at one location of a network, (2) identifies information at another location of the network, and (3) has visual characteristics,
receiving a request from a user to navigate to a different location of the network;
ranking the plurality of links displayed in the first document based on a plurality of ranking values, each ranking value being based on a comparison of a given link of the plurality of links with a plurality of prior links, where the plurality of prior links were displayed to the user in documents from locations of the network different from the first location;
selecting a second location identified by a first link of the plurality of links displayed in the first document based on a first ranking value of the plurality of ranking values associated with the first link; and
receiving and displaying information from the second location,
wherein ranking the links further comprises (i) ranking the given link higher if visual characteristics of the given link correspond to visual characteristics of prior links of the plurality of prior links that were selected by the user, and (ii) ranking the given link lower if the visual characteristics of the given link correspond to prior links of the plurality of prior links that were not selected by the user, wherein a highest-ranked link is selected.
2. The method of claim 1 wherein the comparison is based on a correspondence between the visual characteristics of the given link and the visual characteristics of the plurality of prior links without regards to the different location of the network.
3. The method of claim 1 wherein the visual characteristics comprise text displayed by a link.
4. The method of claim 1 wherein each ranking value is further based on at least one of a timing of selections of prior links, a frequency of selections of prior links, and a pattern of selection of prior links.
5. The method of claim 4 , wherein the ranking value is lowered when a prior link of the plurality of prior links was previously backtracked from.
6. The method of claim 1 , wherein the method further comprises:
receiving a signal from a user input device that is associated with a navigation command, where a navigation command comprises a command to display a different document;
selecting, in response to the navigation command, a link displayed in a second portion of the first document based on a time period determined for a first portion of the first document; and
displaying, on an electronic display, the different document targeted by the selected link.
7. A system comprising one or more processors, the one or more processors configured to:
provide for display to a user a first document at a first location on a network, the first document containing a plurality of links to other documents at other locations of the network where each link in the plurality of links is (1) displayed in a document at one location of a network, (2) identifies information at another location of the network, and (3) has visual characteristics,
receive a request from a user to navigate to a different location of the network;
rank the plurality of links displayed in the first document based on a plurality of ranking values, each ranking value being based on a comparison of a given link of the plurality of links with a plurality of prior links, where the plurality of prior links were displayed to the user in documents from locations of the network different from the first location;
select a second location identified by a first link of the plurality of links displayed in the first document based on a first ranking value of the plurality of ranking values associated with the first link; and
receive and displaying information from the second location,
wherein ranking the links further comprises (i) ranking the given link higher if visual characteristics of the given link correspond to visual characteristics of prior links of the plurality of prior links that were selected by the user, and (ii) ranking the given link lower if the visual characteristics of the given link correspond to prior links of the plurality of prior links that were not selected by the user, wherein a highest-ranked link is selected.
8. The system of claim 7 , wherein the comparison is based on a correspondence between the visual characteristics of the given link and the visual characteristics of the plurality of prior links without regards to the different location of the network.
9. The system of claim 7 , wherein the visual characteristics comprise text displayed by a link.
10. The system of claim 7 , wherein each ranking value is further based on at least one of a timing of selections of prior links, a frequency of selections of prior links, and a pattern of selection of prior links.
11. The system of claim 10 , wherein the ranking value is lowered when a prior link of the plurality of prior links was previously backtracked from.
12. The system of claim 7 , wherein the one or more processors is further configured to:
receive a signal from a user input device that is associated with a navigation command, where a navigation command comprises a command to display a different document;
select, in response to the navigation command, a link displayed in a second portion of the first document based on a time period determined for a first portion of the first document; and
display, on an electronic display, the different document targeted by the selected link.
13. A non-transitory computer-readable medium comprising instructions that, when executed, cause one or more processors to perform a method, the method comprising:
providing for display to a user a first document at a first location on a network, the first document containing a plurality of links to other documents at other locations of the network where each link in the plurality of links is (1) displayed in a document at one location of a network, (2) identifies information at another location of the network, and (3) has visual characteristics,
receiving a request from a user to navigate to a different location of the network;
ranking the plurality of links displayed in the first document based on a plurality of ranking values, each ranking value being based on a comparison of a given link of the plurality of links with a plurality of prior links, where the plurality of prior links were displayed to the user in documents from locations of the network different from the first location;
selecting a second location identified by a first link of the plurality of links displayed in the first document based on a first ranking value of the plurality of ranking values associated with the first link; and
receiving and displaying information from the second location,
wherein ranking the links further comprises (i) ranking the given link higher if visual characteristics of the given link correspond to visual characteristics of prior links of the plurality of prior links that were selected by the user, and (ii) ranking the given link lower if the visual characteristics of the given link correspond to prior links of the plurality of prior links that were not selected by the user, wherein a highest-ranked link is selected.
14. The medium of claim 13 , wherein the comparison is based on a correspondence between the visual characteristics of the given link and the visual characteristics of the plurality of prior links without regards to the different location of the network.
15. The medium of claim 13 , wherein the visual characteristics comprise text displayed by a link.
16. The medium of claim 13 , wherein each ranking value is further based on at least one of a timing of selections of prior links, a frequency of selections of prior links, and a pattern of selection of prior links.
17. The medium of claim 16 , wherein the ranking value is lowered when a prior link of the plurality of prior links was previously backtracked from.
18. The method of claim 1 , wherein the visual characteristic comprise one or more images displayed by a link.
19. The system of claim 7 , wherein the visual characteristic comprise one or more images displayed by a link.
20. The medium of claim 13 , wherein the visual characteristic comprise one or more images displayed by a link.