METHOD FOR FACILITATING SOCIAL NETWORKING BASED ON FASHION-RELATED INFORMATION

• US 20100005105A1
• Filed: 07/02/2008
• Published: 01/07/2010
• Est. Priority Date: 07/02/2008
• Status: Active Grant

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Palo Alto Research Center, Inc. (Xerox Holdings Corp.)
• Inventors: Zhang, Wei, Matsumoto, Takashi, Chu, Maurice K., Begole, James M.A.
• Granted Patent: US 8,386,486 B2
• Time in Patent Office: Days
• US Class Current: 707/100
• CPC Class Codes: G06F 16/90 Details of database functions independent of the retrieved data types
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

One embodiment of the present invention provides a system for facilitating social networking based on fashion-related information. During operation, the system receives fashion-related information from a user. Next, the system extracts the user'"'"'s fashion preferences from the received information and compares the user'"'"'s fashion preference with other users'"'"' fashion preferences. Finally, the system groups users based on similarity of their fashion preferences.

Claims

1. A computer executed method for facilitating social networking based on fashion-related information, comprising:

receiving fashion-related information from a user;

extracting the user'"'"'s fashion preferences from the received information;

comparing the user'"'"'s fashion preference with other users'"'"' fashion preferences; and

grouping users based on similarity of their fashion preferences.

2. The method of claim 1, wherein receiving fashion-related information further comprises receiving an image of the user wearing one or more clothing items.

3. The method of claim 1, wherein extracting the user'"'"'s fashion preferences further comprises extracting a vector of attributes from the fashion-related information.

4. The method of claim 3, wherein the vector of attributes comprise one or more of:

color, cut, style, size, material, fabric, texture, thickness, buttons, brand, function, comfort, pattern, weave, category and subcategories, collection, designer, price, fashion line, attractiveness, fashionableness, fit, durability, quality, symmetry, layering, country of origin, ease of care, laces, shine, user-defined tags, user demographics, a similarity measure to other users, user actions, and a classifier which captures a user'"'"'s preferences.

5. The method of claim 1, wherein grouping users comprises:

clustering users into groups based on a similarity measure between two user'"'"'s fashion preferences;

identifying from the groups a group which is associated with the user; and

ranking the other users in the group based on the user'"'"'s fashion preferences.

6. The method of claim 5, wherein clustering users into groups further comprises:

constructing a graph of users, wherein a cost of a link in the graph between a first user and a second user is inversely proportional to the similarity measure between the fashion preferences of the first user and the second user; and

partitioning the graph of users into disjoint groups using a minimum-cost graph-cut approach, such that fashion preferences of a user within one group have a high similarity measure to fashion preferences of a user within the same group and a low similarly measure to fashion preferences of a user within a different group.

7. The method of claim 5, further comprising:

presenting to the user the ranked other users in the same group as the user;

receiving user feedback from the user; and

updating the user'"'"'s fashion preferences based on the received feedback.

8. The method of claim 1, further comprising recommending items to the user based on the user'"'"'s fashion preferences and the fashion preferences of other users.

9. An apparatus for facilitating social networking based on fashion-related information, comprising:

a receiving mechanism configured to receive fashion-related information from a user;

an extraction mechanism configured to extract the user'"'"'s fashion preferences from the received information;

a comparison mechanism configured to compare the user'"'"'s fashion preference with other users'"'"' fashion preferences; and

a grouping mechanism configured to group users based on similarity of their fashion preferences.

10. The apparatus of claim 9, wherein while receiving fashion-related information, the receiving mechanism is further configured to receive an image of the user wearing one or more clothing items.

11. The apparatus of claim 9, wherein while extracting the user'"'"'s fashion preferences, the extraction mechanism is further configured to extract a vector of attributes from the fashion-related information.

12. The apparatus of claim 11, wherein the vector of attributes comprise one or more of color, cut, style, size, material, fabric, texture, thickness, buttons, brand, function, comfort, pattern, weave, category and subcategories, collection, designer, price, fashion line, attractiveness, fashionableness, fit, durability, quality, symmetry, layering, country of origin, ease of care, laces, shine, user-defined tags, user demographics, a similarity measure to other users, user actions, a classifier which captures a user'"'"'s preferences, and other attributes of clothing.

13. The apparatus of claim 9, wherein while grouping users, the grouping mechanism is further configured to:

cluster users into groups based on a similarity measure between two user'"'"'s fashion preferences;

identify a group which is associated with the user from the groups; and

rank the other users in the group based on the user'"'"'s fashion preferences.

14. The apparatus of claim 13, wherein while clustering users into groups, the clustering mechanism is further configured to:

construct a graph of users, wherein a cost of a link in the graph between a first user and a second user is inversely proportional to the similarity measure between the fashion preferences of the first user and the second user; and

partition the graph of users into disjoint groups using a minimum-cost graph-cut approach, such that fashion preferences of a user within one group have a high similarity measure to fashion preferences of a user within the same group and a low similarly measure to fashion preferences of a user within a different group.

15. The apparatus of claim 9, further configured to:

present ranked users to the user;

receive user feedback; and

update the user'"'"'s fashion preferences based on the received feedback.

16. The apparatus of claim 9, further configured to recommend items to the user based on the user'"'"'s fashion preferences and the fashion preferences of other users.

17. A computer-readable storage device storing instructions that when executed by a computer cause the computer to perform a method for facilitating social networking based on fashion-related information, the method comprising:

receiving fashion-related information from a user;

extracting the user'"'"'s fashion preferences from the received information;

comparing the user'"'"'s fashion preference with other users'"'"' fashion preferences; and

grouping users based on similarity of their fashion preferences.

18. The storage device of claim 17, wherein receiving fashion-related information further comprises receiving an image of the user wearing one or more clothing items.

19. The storage device of claim 17, wherein extracting the user'"'"'s fashion preferences further comprises extracting a vector of attributes from the fashion-related information.

20. The storage device of claim 19, wherein the vector of attributes comprise one or more of color, cut, style, size, material, fabric, texture, thickness, buttons, brand, function, comfort, pattern, weave, category and subcategories, collection, designer, price, fashion line, attractiveness, fashionableness, fit, durability, quality, symmetry, layering, country of origin, ease of care, laces, shine, user-defined tags, user demographics, a similarity measure to other users, user actions, a classifier which captures a user'"'"'s preferences, and other attributes of clothing.

21. The storage device of claim 17, wherein grouping users further comprises:

clustering users into groups based on a similarity measure between two user'"'"'s fashion preferences;

identifying a group which is associated with the user from the groups; and

ranking the other users in the group based on the user'"'"'s fashion preferences.

22. The storage device of claim 21, wherein clustering users into groups further comprises:

constructing a graph of users, wherein a cost of a link in the graph between a first user and a second user is inversely proportional to the similarity measure between the fashion preferences of the first user and the second user; and

partitioning the graph of users into disjoint groups using a minimum-cost graph-cut approach, such that fashion preferences of a user within one group have a high similarity measure to fashion preferences of a user within the same group and a low similarly measure to fashion preferences of a user within a different group.

23. The storage device of claim 17, further comprising:

presenting ranked users to the user;

receiving user feedback; and

updating the user'"'"'s fashion preferences based on the received feedback.

24. The method of claim 17, further comprising recommending items to the user based on the user'"'"'s fashion preferences and the fashion preferences of other users.