Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Method and System for Detecting Communities in a Social Network based on Images

IP.com Disclosure Number: IPCOM000218121D
Publication Date: 2012-May-22
Document File: 2 page(s) / 16K

Publishing Venue

The IP.com Prior Art Database

Related People

Jia Li: INVENTOR [+4]

Abstract

A method and system for detecting communities in a social network based on images is disclosed. The method and system constructs the social network by identifying images of different users from a common image repository. The social network is then used to automatically infer communities among the users. The social network can serve as a knowledge base for social recommendation and propagation based on uniqueness of related content within each community of the social network.

This text was extracted from a Microsoft Word document.
This is the abbreviated version, containing approximately 52% of the total text.

Method and System for Detecting Communities in a Social Network based on Images

Abstract

A method and system for detecting communities in a social network based on images is disclosed.  The method and system constructs the social network by identifying images of different users from a common image repository.  The social network is then used to automatically infer communities among the users.  The social network can serve as a knowledge base for social recommendation and propagation based on uniqueness of related content within each community of the social network.

Description

Disclosed is a method and system for detecting communities in a social network based on images.  The method and system constructs the social network by identifying images of or containing different users from a common image repository.  The social network is used to represent the relationship and interaction between one or more users.  The social network is constructed based on availability of different images corresponding to multiple users in the common image repository.

Thereafter, factors such as, but not limited to, online behavior, user images and tags of a group of users are analyzed to determine the group community.  The group community is determined in such a way that similar user images and tags are allocated to same community and different user images and tags are allocated to different communities.  An algorithm is used to analyze the user images and tags to allocate them either the same co...