The Prior Art Database and Publishing service will be updated on Sunday, February 25th, from 1-3pm ET. You may experience brief service interruptions during that time.
Browse Prior Art Database

Social Tagging Utilizing Data Mining

IP.com Disclosure Number: IPCOM000236912D
Publication Date: 2014-May-21
Document File: 2 page(s) / 133K

Publishing Venue

The IP.com Prior Art Database


Disclosed are a system and method to incorporate metadata associated with a photograph with other data sources in an algorithm in order to increase the accuracy of subject identification in a photograph for social media applications.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 2

Social Tagging Utilizing Data Mining

Current systems for social tagging (i.e. via social networking applications) often mislabel posts, photographs, and videos that have similar content to another post, photograph, or video, because such systems are strictly based on a single data source.

A method and system is needed to increase the confidence and accuracy of social


The novel contribution is a system and method to increase the accuracy of identifying an individual(s) in a photograph for social tagging. The system and method incorporate the photo's metadata with other data sources that increase the confidence and accuracy of subject identification in a photograph. Those data sources include:

• Social media posts/messages/communications: the system compares posts/messages/check-ins

• Location of person: determined by location of credit card purchases, phone location, system access, etc.

• Contents of preceding/proceeding photo including comparison of people/known associates

• Unique physical traits such as scars, marks, and wrinkles

The system and method improve the accuracy of social tagging in photos by including external data sources into the algorithm.

Example Embodiment

When reviewing photographs under the traditional facial recognition systems, a photograph of a child can be confused with that of a parent because the facial characteristics are very similar. The system and method described herein can increase the accuracy by using additional data taken from a photograph when compared against external data sources (i.e. social media, location, etc.). Examples of criteria that can be considered include, but are not limited to: age projection, what

the person is weari...