Statistical Device Graph Model for Cross-Device Consumer Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately understand consumer interactions with online resources across multiple devices, particularly when consumers share devices or use different login identifiers, leading to inaccurate analytics and difficulty in linking device interactions to specific consumers.
Innovation Solution
A statistical device graph model is generated based on observational data from multiple data providers, which includes device and login identifiers, to infer whether a device is private, shared, or public, and to link interactions across devices used by a single consumer, providing insights into device usage patterns and confidence levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional cross-device visitor identification techniques are used, then device interactions can be tracked, but accuracy in identifying private devices and linking interactions to specific consumers deteriorates when consumers share devices or use different login identifiers
Solution Approach 1:
The patent introduces a statistical device graph model as an intermediary representation that indirectly identifies consumer-device relationships. Instead of directly tracking consumers across devices, the model uses observational data from multiple data providers to infer device types (private, shared, public) and establish probabilistic links between devices and consumers, thereby resolving the contradiction between identification accuracy and adaptability to diverse usage scenarios
Solution Approach 2:
The patent changes the parameter of device identification from deterministic (direct consumer ID matching) to probabilistic (statistical inference based on observational patterns). By analyzing login identifiers, device identifiers, and interaction patterns across multiple data providers, the system dynamically adjusts confidence levels and device type classifications, enabling accurate identification even when consumers use different login identifiers or share devices
2Loss of information
If observational data from multiple data providers is collected and analyzed, then insights into device usage patterns and consumer interactions are improved, but system complexity and computational requirements worsen
Solution Approach 1:
The patent segments the complex task of cross-device identification into manageable components: collecting observational data from multiple data providers, building a statistical device graph model, inferring device types, and establishing device-linkage relationships. This segmentation allows the system to handle large volumes of observational data from multiple sources without overwhelming computational complexity, as each component processes specific aspects of the data independently
Solution Approach 2:
The statistical device graph model serves multiple functions simultaneously: it tracks consumer interactions across devices, infers device types (private, shared, public), establishes device-linkage relationships, and provides confidence levels for identifications. This multi-functionality reduces overall system complexity by consolidating multiple analytical tasks into a single unified model rather than requiring separate systems for each function
Data Source
AI summary
An analytics server receives data characterizing consumer interactions that are observed by a cross-section of data providers, which may include, for example, website administrators, campaign managers, application developers, and the like. Such observational data includes device and login identifiers for a particular interaction, and optionally, timestamp information indicating when the interaction occurred. A statistical device graph model is generated based on this observational data. The statistical device graph model allows inferences to be drawn with respect to whether a given device is a private device, a shared device, or a public device. This, in turn, allows private devices which are “owned” by a single consumer to be identified. Depending on the type of observational data collected by the data providers, a wide range of additional insights can be drawn from the statistical device graph model, including for example, device usage patterns and confidence levels.