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

Method and System for Targeting Advertisements based on Pattern Analysis of Social Media Images

IP.com Disclosure Number: IPCOM000240511D
Publication Date: 2015-Feb-04
Document File: 3 page(s) / 94K

Publishing Venue

The IP.com Prior Art Database


A method and system is disclosed for targeting advertisements based on pattern analysis of social media images.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 44% of the total text.

Page 01 of 3

Method and System for Targeting Advertisements based on Pattern Analysis of Social Media Images

On social networking sites, users can upload images. Usually, one or more objects shown in such images are analyzed and appropriate advertisements based on the one or more objects are displayed. For example, a social network user posts an image of the user where the user is carrying a hand bag in the image. The image is analyzed and advertisements for similar hand bags are rendered to the user. However, such advertisements may not be displayed at the right time to the user. Perhaps, the user might have recently purchased the hand bag and is no longer interested in buying a new hand bag.

Disclosed is a method and system for targeting advertisements based on pattern analysis of social media images. The method and system performs the pattern analysis of the social media images to predict a next purchase decision of a user. Based on an outcome of the pattern analysis, targeted advertisements are rendered to the user. For example, using the pattern analysis, the system can determine that a user is likely to buy a new hand bag after 6 months or whenever a newer model gets released. Thus, an advertisement is displayed to the user only after 6 months or when a newer model is released.

The method includes analysis of images on social networks sites to determine people present in an image and objects associated with each person. This analysis is stored in a central repository and a pattern analysis is performed. The pattern analysis determines trends such as the historical frequency of user purchases and item replacement. The results of the pattern analysis are used to determine items which a user is most likely to purchase currently and to display advertisements related to such items.

The method and system recognizes objects in social media images to identify objects within the image. After recognition, the method and system tags identified objects with
a date, location and a person in possession of an object. The tagging is carried out to analyze metadata of the object with other images containing the same object. Based on the estimated age of the object, the system identifies objects in a recent image that the user may now wish to replace. An appropriate advertisement for replacement objects can be shown alongside the image. Alternatively, the identified objects are analyzed to determine if a newer model is available. If a newer model is identified, an appropriate advertisement for the newer model can be shown alongside the image.

In addition, the system analyzes the central repository to extract demographics information. By aggregating an identified object across multiple social media images, a creator of a product can build an understanding of the demographics of people who are using products of the creator. The understanding of the demographics can include determining the age, gender, location, and so forth of the people pictured using the...