Cross-Device Identity Clustering Beyond Dynamic IP Addresses
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods fail to reliably identify associations between different devices used by the same user across multiple Internet-connected devices due to transient and dynamic IP addresses and anonymous communication, making it difficult for commercial entities to effectively target users across their devices.
Innovation Solution
A data processing system that utilizes a pairing engine to identify candidate device pairs based on device activity history, generates feature vectors using a feature vector generation engine, scores these pairs with a scoring engine, and clusters them using a clustering engine to form digital identity groups, enabling efficient association of devices based on their usage patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional IP address-based identification methods are used to track users across devices, then user identification can be implemented with simple infrastructure, but the reliability of user identification deteriorates due to dynamic and transient IP addresses
Solution Approach 1:
The patent introduces an intermediary identification system that uses device fingerprints, cookies, and behavioral patterns as mediators between the user and the identification system. Instead of relying directly on transient IP addresses, the system uses these intermediary markers that persist across sessions and can be correlated to identify the same user across multiple devices and time periods.
Solution Approach 2:
The patent replaces the mechanical IP address-based identification system with a software-based identification system that uses device fingerprints, browser characteristics, and behavioral analysis. This substitution allows for more reliable user identification that is not dependent on network infrastructure or IP address assignment.
2Measurement precision
If device associations are tracked using detailed activity data to improve identification accuracy, then user identification precision improves, but data processing requirements and computational resources increase
Solution Approach 1:
The patent performs preliminary actions by pre-computing and storing device fingerprints, behavioral baselines, and association rules before they are needed for identification. Device characteristics are fingerprinted in advance, and behavioral patterns are established through initial observation periods, allowing for faster and more efficient real-time identification without intensive computational resources during actual user tracking.
Solution Approach 2:
The patent applies partial action by selectively processing only the most relevant features and data points for device association. Instead of analyzing all possible device attributes equally, the system focuses on key identifying characteristics and uses probabilistic methods to process data efficiently, achieving high precision without requiring exhaustive computation of all available data.
3Adaptability or versatility
If comprehensive device activity data is collected to improve user profiling, then the ability to reach users across devices improves, but user privacy concerns and data security requirements increase
Solution Approach 1:
The patent extracts and separates personally identifiable information (PII) from device activity data. The system processes and analyzes device behavior patterns without storing or transmitting actual user identities, personal information, or sensitive data. This extraction of harmful or sensitive elements allows for effective cross-device tracking while minimizing privacy risks and security requirements.
Solution Approach 2:
The patent transforms raw device activity data into anonymized behavioral parameters and aggregated statistics. By changing the form and representation of data from identifiable detailed records to anonymized pattern recognition data, the system maintains its ability to profile user behavior and enable cross-device communication while significantly reducing privacy concerns and security vulnerabilities.
Data Source
AI summary
A data processing system performs data processing of raw or preprocessed data. In some embodiments, the data processing system includes a connectivity overlay engine comprising a data ingester, a connectivity generator, an event access control system, and a feature vector generation framework.