Cross-Platform User Identity Matching via Image and Text Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users registered across multiple online systems often provide different information, leading to inefficiencies in content delivery as each system lacks a comprehensive understanding of the user, resulting in poorly tailored content due to inconsistent profile data and lack of cross-system identification.
Innovation Solution
An online system matches users across multiple platforms using image and textual data comparison, employing deep learning models to generate similarity scores and predict identity matches, thereby building a comprehensive user profile for improved content delivery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If users provide different profile information on each online system, then each system can deliver content tailored to user preferences, but the system lacks comprehensive user understanding leading to poor content delivery
Solution Approach 1:
The patent merges user profile information from multiple online systems by using image recognition to identify the same user across different platforms. The system combines textual profile data and image data into a unified user identity, allowing comprehensive user understanding while maintaining the ability to deliver tailored content across all systems.
Solution Approach 2:
The patent introduces an intermediary matching system that uses image recognition technology as a mediator between different online systems. This intermediary layer connects user profiles across systems by comparing images and textual data, enabling information sharing without requiring direct integration between each system.
2Adaptability or versatility
If users use different profile photos on each online system, then users can present themselves differently on each platform, but cross-system identification fails leading to inefficiencies
Solution Approach 1:
The patent merges image data from different online systems using deep learning models that can recognize the same user across varying photo conditions. The system combines image similarity metrics with textual profile data to reliably identify users even when they use different profile photos on different platforms.
Solution Approach 2:
The patent applies parameter changes by using deep learning models that can adapt to different image characteristics, lighting conditions, and photo styles. The system adjusts its recognition parameters to accommodate diverse profile photos while maintaining consistent user identification across systems.
3Device complexity
If each online system maintains separate user profiles, then system independence is preserved, but content delivery becomes poorly tailored due to lack of comprehensive user data
Solution Approach 1:
The patent introduces an intermediary matching service that operates between independent online systems. This intermediary uses image recognition and data comparison to create user identity mappings across systems, allowing each system to maintain its independence while sharing comprehensive user profile information for improved content delivery.
Solution Approach 2:
The patent creates a universal user identification framework that can be applied across multiple online systems. The system provides multi-functional capabilities including user matching, profile consolidation, and content personalization, serving multiple purposes without requiring changes to individual system architectures.
Data Source
AI summary
An online system matches a user across multiple online systems based on image data for the user (e.g., profile photo) regardless whether the image data is from the online system, a different but related online system or a third party system. For example, to match the user across a social networking system and INSTAGRAM™ system, the online system compares the similarity between images of the user from both systems in addition to similarity of textual information in the user profiles on both systems. The similarity of image data and the similarity of textual information associated with the user are used by the online system as indicators of matched user accounts belonging to the same user across both systems. The online system applies models trained using deep learning techniques to match a user across multiple online systems based on the image data and textual information associated with the user.