Image-Based Ad Targeting via Automated Visual Content Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing advertising methods on social networking sites fail to effectively target users' interests, as profiles filled by users may not accurately represent their actual interests, leading to irrelevant advertisements and inefficient traffic generation for sponsors.
Innovation Solution
A system and method that analyzes users' images to generate personalized advertisements by categorizing and comparing image content, using a classifier to select advertising images based on user profiles, combining these images with sponsor content to create tailored advertisements displayed on web pages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If advertisements are targeted based on user profiles filled by users, then advertising relevance may be improved, but the accuracy of interest representation deteriorates because user-filled profiles do not always accurately portray their interests
Solution Approach 1:
The patent replaces manual user profile filling with automated image-based interest detection. Instead of relying on users to accurately describe their interests through text forms, the system automatically analyzes images users have uploaded to the social network to infer their interests. This substitution of mechanical manual input with automated computer vision analysis resolves the contradiction by eliminating the inaccuracy inherent in user self-description while maintaining high relevance through objective image content analysis.
Solution Approach 2:
The system enables self-service interest profiling by automatically analyzing users' own uploaded images to generate interest profiles without requiring user input. The images themselves serve as the data source, allowing the system to serve itself in extracting interest information that users may not be able or willing to provide manually, thus improving both accuracy and relevance simultaneously.
2Ease of manufacture
If user profiles are generated manually by users, then profile creation is simple, but the profiling accuracy deteriorates because users may not accurately represent their true interests
Solution Approach 1:
The patent substitutes manual profile creation with automated image analysis. Instead of requiring users to fill out forms or provide demographic information, the system automatically processes images uploaded to the social network using computer vision algorithms. This maintains ease of use (users simply upload images as they normally would) while dramatically improving profile accuracy through objective visual content analysis that reflects actual user behavior and interests.
Solution Approach 2:
The system performs self-service profiling by automatically extracting interest information from users' uploaded images without requiring any active participation from users in the profiling process itself. The images that users naturally upload to share moments and interests become the raw material for accurate profile generation, eliminating the need for users to manually describe their interests while maintaining high accuracy.
3Ease of manufacture
If advertisements are displayed based on keywords on web pages, then advertising deployment is easy, but advertising relevance to user interests deteriorates
Solution Approach 1:
The patent replaces keyword-based advertising with image-based advertising. Instead of relying on text keywords that may not reflect user interests, the system analyzes visual content of images uploaded by users to determine their interests. This substitution maintains the ease of automated advertising deployment while significantly improving relevance, as image analysis directly captures what users are actually engaged with rather than relying on potentially inaccurate textual keywords.
Solution Approach 2:
The system introduces image analysis as an intermediary between user behavior and advertising selection. Rather than directly matching keywords to ads or relying on self-reported user preferences, the system uses automated image content analysis as an intermediary to infer user interests from the visual content they share. This intermediary layer provides more accurate interest representation while maintaining automated deployment through computer vision processing.
Data Source
AI summary
An advertising system and method are disclosed for generating advertising content based on a user's images. These can be photographs owned by a user of a web page. An advertising template for an advertising sponsor is provided. For a user viewing the web page, a set of the user's images is categorized, based on image content of the images in the set. The categorization is performed with at least one classifier which has been trained on a finite set of image categories. An advertising image is selected for advertising content based on the categorization of the user's images. The template is combined with the selected advertising image to form the advertising content. The advertising content can then be displayed to the user on the web page. A user profile is also disclosed which can be used as the categorized user images.