Ranking and suggesting candidate objects

• US 7,685,200 B2
• Filed: 03/01/2007
• Issued: 03/23/2010
• Est. Priority Date: 03/01/2007
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Gunawardena, Dinan, Marbach, Peter, Vojnović, Milan, Cruise, Richard James
• Primary Examiner(s): Abel-Jalil; Neveen
• Assistant Examiner(s): KINSAUL, DANIEL W
• Application Number:
  US11/681,031
• Publication Number:
  US 20080215583A1
• Time in Patent Office: 1,118 Days
• Field of Search: 707/7
• US Class Current: 707/748
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

A method of ranking and suggesting objects, such as tags, is described in which a group of objects are suggested to a user in relation to a particular information object, such as a web page or file. A user either selects one of the suggestions or enters an alternative object and on receipt of data identifying the particular object selected by the user, the group of suggestions is updated based on whether the object selected by the user was one of the suggestions or not. The updating of the group may also be based on other factors, such as ranking parameters associated with each of the objects.

Claims

1. A method of ranking and suggesting tags implemented by a processor executing computer-executable instructions contained in one or more computer-readable storage media, the method comprising:

proposing, by the processor, a set of suggested tags to a user;

receiving, by the processor, data identifying a tag selected by a user; and

updating, by the processor, said set of suggested tags on the basis of at least whether said selected tag is a member of said set of suggested tags;

wherein updating said set of suggested tags on the basis of at least whether said selected tag is a member of said set of suggested tags comprises;

determining if said selected tag is a member of said set of suggested tags;

if said selected tag is a member of said set, maintaining said set of suggested tags unchanged; and

if said selected tag is not a member of said set, updating said set of suggested tags,

wherein updating said set of suggested tags comprises;

increasing a ranking parameter associated with said selected tag and decreasing a ranking parameter associated with each of said plurality of tags which is neither said selected tag nor a member of said set of suggested tags; and

updating said set of suggested tags based on said ranking parameters;

wherein the updating said set of suggested tags on the basis of at least whether said selected tag is a member of said set of suggested tags further comprises;

updating said set of suggested tags on the basis of whether said selected tag is a member of said set of suggested tags and whether a randomly generated variable exceeds a defined threshold; and

wherein the randomly-generated variable is generated when data identifying a tag selected by the user is received, wherein the updating occurs only when the value of the variable exceeds the threshold, wherein tag reinforcement resulting from suggesting tags in the suggestion set is mitigated.

2. A method according to claim 1, wherein said set of suggested tags are selected from a plurality of tags associated with an information object.

3. A method according to claim 2, wherein the random variable is uniformly distributed over a predetermined range and said threshold is statically set in the middle of the predetermined range.

4. A method according to claim 3, wherein said information object comprises one of a uniform resource locator (URL), a web page, a file, an image, an audio clip and a video clip.

5. A method according to claim 1, wherein the proposing the set of suggested tags to a user comprises:

accessing said set of suggested tags; and

displaying said set of suggested tags to said user.

6. A method according to claim 1, wherein the value of the threshold is learned dynamically based upon a weighted moving exponential average of the number of user entered tags at any given tagging.

7. A method according to claim 1, wherein updating said set of suggested tags on the basis of at least whether said selected tag is a member of said set of suggested tags comprises:

updating said set of suggested tags on the basis of whether said selected tag is a member of said set of suggested tags and whether said selected tag meets eligibility criteria.

8. A method according to claim 1, wherein updating said set of suggested tags comprises:

removing one of said set from said set of suggested tags, wherein said one is selected substantially at random from said set of suggested tags; and

adding said selected tag to said set of suggested tags.

9. A method according to claim 1, wherein updating said set of suggested tags comprises:

determining if said selected tag meets eligibility criteria; and

if said selected tag meets said eligibility criteria, removing one of said set from said set of suggested tags, wherein said one is selected substantially at random from said set of suggested tags; and

adding said selected tag to said set of suggested tags.

10. A method according to claim 9, wherein said eligibility criteria comprises:

a vote count parameter associated with said selected tag exceeds a vote count parameter associated with at least one member of said set of suggested tags.

11. A method according to claim 10, further comprising:

if said selected tag is not a member of said set, incrementing said vote count parameter associated with said selected tag.

12. A method according to claim 1, wherein updating said set of suggested tags based on said ranking parameters comprises:

updating said set of suggested tags to comprise a subset of said plurality of tags having largest ranking parameters.

13. A method according to claim 1, wherein selecting said set of suggested tags based on said ranking parameters further comprises:

selecting said set of suggested tags by selecting a random set of tags from said plurality of tags, wherein a probability of selection of a tag from said plurality of tags to create the random set of tags is proportional to a ranking parameter associated with each said tag of the plurality of tags.

14. A system for ranking and suggesting tags for uniform resource locators (URLs), the system comprising:

a server in communication with a client, the server including a first processor executing computer-executable instructions contained in one or more computer readable media for implementing;

a tag suggestion module for selecting a random second plurality of tags from a first plurality of tags associated with an information object, wherein a probability of selection of a tag as one of the random second plurality of tags is proportional to a ranking parameter associated with each said tag, wherein the tag suggestion module proposes said second plurality of tags to a user at the client as a set of suggested tags;

a tag entry server side component for receiving from the client data identifying a tag selected by the user at the client, and for updating at least one ranking parameter associated with at least one of the first plurality of tags based on a ranking rule;

wherein said updating at least one ranking parameter associated with at least one of the first plurality of tags further comprises;

increasing at least one ranking parameter associated with said tag selected by the user and decreasing at least one ranking parameter associated with each of said first plurality of tags which is neither said tag selected by the user nor a member of said second plurality of tags; and

updating said second plurality of tags based on said ranking parameters.

15. A system according to claim 14, further comprising a search module implemented by the server for searching tags by the user of the client based on rankings obtained from a plurality of users of the system.

16. A system according to claim 14, wherein said ranking rule comprises:

increasing the value of a ranking parameter associated with said tag selected by said user; and

decreasing the value of a ranking parameter associated with each of the plurality of tags which is not said tag selected by said user.

17. A system according to claim 14, wherein said ranking rule comprises, if said tag selected by said user is not one of said second plurality of tags:

increasing the value of a ranking parameter associated with said selected tag; and

decreasing the value of a ranking parameter associated with each of the first plurality of tags which is neither said tag selected by said user nor one of said second plurality of tags.

18. A method of ranking and suggesting tags for uniform resource locators (URLs), the method comprising:

receiving notification of starting a tagging instance by a user at a client device for a currently-navigated URL, the client device including a client processor coupled to a client memory storing client-executable instructions for initiating a tagging interface at the client, wherein the tagging interface comprises an add-in for a web browser;

identifying by a server in communication with the client a suggestion set of tags corresponding to the currently navigated URL, the server including a server processor coupled to a server memory storing server-executable instructions for implementing a tag suggestion module for selecting the suggestion set of tags as a random second plurality of tags selected from a first plurality of tags associated with the currently-navigated URL based on ranking parameters associated with each of the first plurality of tags,

selecting, by the tag suggestion module, the random second plurality of tags from the first plurality of tags associated with the URL, wherein a probability of selection of a tag as one of the random second plurality of tags is proportional to a ranking parameter associated with each said tag;

proposing, by the tag suggestion module said random second plurality of tags to the user at the client as the suggestion set of tags, wherein the client presents a tag entry screen on the client, the tag entry screen including a tag entry box for the user to enter one or more tags deemed relevant by the user to the currently navigated URL;

receiving data identifying a selected tag selected by a user via the tag entry screen; and

updating said first plurality of tags on a basis of whether said selected tag is a member of said suggestion set of tags and whether a randomly generated variable exceeds a defined threshold, wherein the randomly-generated variable is generated for each time the user selects a tag, wherein the updating occurs only when the value of variable exceeds the threshold so that tag reinforcement resulting from suggesting tags in the suggestion set is mitigated; and

when said tag selected by said user is not one of said second plurality of tags and the randomly generated variable exceeds the threshold;

increasing the value of a ranking parameter associated with said selected tag; and

decreasing the value of a ranking parameter associated with each of the first plurality of tags which is neither said tag selected by said user nor one of said second plurality of tags.