System and method for user classification based on social network aware content analysis
First Claim
1. A non-transitory computer-readable medium comprising computer instructions embodied therein, wherein the computer instructions, when executed by at least one computing device, configure the at least one computing device to perform actions comprising:
- selecting a user from a plurality of users in a social network of a target user;
scanning metadata associated with content items shared by the selected user;
determining a list of keywords associated with the content items based on the metadata;
accessing social network data associated with the selected user, the social network data including a relationship classifier that classifies a relationship between the selected user and the target user and a social network distance between the target user and the selected user, and a trust classifier that classifies a trust level between the selected user and the target user;
determining a plurality of keyword scores based on the social network data, each of the plurality of keyword scores corresponding to one of the keywords on the list of keywords, wherein each of the plurality of scores is weighted by the trust classifier and an adjustment factor inversely proportional to the social network distance;
normalizing individual keyword scores of the plurality of keywords scores based on a number of the content items shared by the selected user in comparison to a number of content items shared by other selected users of the plurality of users when the number of the content items shared by the selected user is greater than the number of content items shared by other selected users of the plurality of users; and
generating a profile for the target user, the profile having the list of keywords and the corresponding plurality of keyword scores based on the scoring of each keyword.
1 Assignment
0 Petitions
Accused Products
Abstract
A target user'"'"'s profile may be generated based on scanning metadata related to various content items made available for sharing by the user. For instance, some content items may be made available for sharing by the target user while other items are made available for sharing by one or more other users in the target user'"'"'s social network. A list of keywords related to each item may be generated from the metadata. Each keyword may then be scored using social network data. For example, social network data corresponding to one or more users associated with each content item may be used to score keywords related to that item. Scoring may be based on factors such as social network distance, relationship type, trust, length of time in the social network, amount of content, amount of metadata, and other factors.
-
Citations
24 Claims
-
1. A non-transitory computer-readable medium comprising computer instructions embodied therein, wherein the computer instructions, when executed by at least one computing device, configure the at least one computing device to perform actions comprising:
-
selecting a user from a plurality of users in a social network of a target user; scanning metadata associated with content items shared by the selected user; determining a list of keywords associated with the content items based on the metadata; accessing social network data associated with the selected user, the social network data including a relationship classifier that classifies a relationship between the selected user and the target user and a social network distance between the target user and the selected user, and a trust classifier that classifies a trust level between the selected user and the target user; determining a plurality of keyword scores based on the social network data, each of the plurality of keyword scores corresponding to one of the keywords on the list of keywords, wherein each of the plurality of scores is weighted by the trust classifier and an adjustment factor inversely proportional to the social network distance; normalizing individual keyword scores of the plurality of keywords scores based on a number of the content items shared by the selected user in comparison to a number of content items shared by other selected users of the plurality of users when the number of the content items shared by the selected user is greater than the number of content items shared by other selected users of the plurality of users; and generating a profile for the target user, the profile having the list of keywords and the corresponding plurality of keyword scores based on the scoring of each keyword. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12)
-
-
13. A computer-implemented method of profiling a target user having a social network including a plurality of users, the method comprising:
-
selecting a user from the plurality of users in the social network of the target user; scanning metadata associated with content items shared by the selected user; determining a list of keywords associated with the content items based on the metadata; accessing social network data associated with the selected user, the social network data including a relationship classifier that classifies a relationship between the selected user and the target user and a social network distance between the target user and the selected user, and a trust classifier that classifies a trust level between the selected user and the target user; determining a plurality of keyword scores based on the social network data, each of the plurality of keyword scores corresponding to one of the keywords on the list of keywords, wherein each of the plurality of scores is weighted by the trust classifier and an adjustment factor inversely proportional to the social network distance; normalizing individual keyword scores of the plurality of keywords scores based on a number of the content items shared by the selected user in comparison to a number of content items shared by other selected users of the plurality of users when the number of the content items shared by the selected user is greater than the number of content items shared by other selected users of the plurality of users; and generating a profile for the target user, the profile having the list of keywords and the corresponding plurality of keyword scores based on the scoring of each keyword. - View Dependent Claims (14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24)
-
Specification