Audience Media Recommendation via Individual Viewer Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing media content recommendation systems fail to effectively recommend content to groups of viewers based on the dynamic traits and interactions of individual viewers, such as emotions and body positions, which are crucial for personalized experiences.
Innovation Solution
A system and method that utilize image processing to capture key attributes of individual viewers, create group profiles, and provide recommendations based on these attributes, including emotions, demographics, and location, using a combination of facial and body analysis, and machine-learning algorithms to predict preferred media content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional recommendation systems are used, then implementation is simple, but recommendation accuracy for groups is poor
Solution Approach 1:
The system segments the audience into individual viewers and analyzes each person's unique traits (emotions, body positions, demographics) separately, then synthesizes these individual analyses into a comprehensive group recommendation. This segmentation approach enables precise measurement of each viewer's characteristics while maintaining overall system functionality.
Solution Approach 2:
The patent introduces image processing technology and machine learning algorithms as intermediary components between the viewers and the recommendation engine. These intermediaries capture and analyze viewer attributes (emotions, body positions) and translate them into actionable recommendation data, bridging the gap between simple system operation and accurate group recommendations.
2Adaptability or versatility
If dynamic viewer traits are analyzed, then recommendation relevance improves, but processing time increases
Solution Approach 1:
The system performs preliminary actions by continuously capturing and pre-processing viewer image data in the background, extracting key attributes (emotions, body positions, demographics) before they are needed for recommendations. This allows the system to have viewer trait data ready when generating recommendations, reducing actual processing time while maintaining high personalization.
Solution Approach 2:
The patent replaces manual or traditional mechanical analysis methods with automated image processing and machine learning algorithms. This substitution enables rapid analysis of dynamic viewer traits (emotions, body positions) in real-time, achieving high adaptability without significant time loss.
3Measurement precision
If image processing is used to capture viewer attributes, then measurement precision improves, but computational load increases
Solution Approach 1:
The system extracts only the essential and most relevant viewer attributes from image data, such as emotions, body positions, and key demographic features. By taking out only these critical elements rather than processing all possible image data, the system achieves high measurement precision for key traits while reducing overall computational energy requirements.
Data Source
AI summary
Aspects of the subject disclosure may include, for example, a device, that has a processing system including a processor; and a memory that stores executable instructions that, when executed by the processing system, facilitate performance of operations, where the operations include detecting each individual of an audience viewing media content on user equipment; retrieving a user profile for each individual of the audience resulting in user profiles; creating a group profile from the user profiles; determining, based on the group profile, a recommendation for viewing a candidate media content; and providing the recommendation to the user equipment for the audience. Other embodiments are disclosed.