Linked-Device Content Selection Through Probabilistic Activity Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing content selection systems struggle to effectively select content for display on devices when network activity is split across multiple computing devices, as they have limited access to comprehensive network activity information.
Innovation Solution
A data processing system links computing devices based on anonymous identifiers, determining a match probability using positive and negative match probabilities and weighting factors, and generates a data structure to indicate linked devices, enabling more accurate content selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If content selection is based on partial network activity from a single device, then content selection can be performed with available data, but the accuracy and relevance of content selection deteriorates due to incomplete network activity information
Solution Approach 1:
The patent merges network activity data from multiple computing devices by creating a unified data structure that aggregates activities across devices. The system identifies relationships between devices and combines their network activities into a consolidated view, enabling more accurate content selection that reflects the complete user behavior pattern across all devices rather than just isolated single-device activity.
2Loss of information
If multiple computing devices are linked together, then comprehensive network activity information becomes available for content selection, but the system complexity increases due to device matching and probability calculations
Solution Approach 1:
The system performs self-service by automatically identifying and linking computing devices through probabilistic matching algorithms. The device linking process occurs autonomously without manual intervention, using calculated match probabilities to determine relationships between devices. This self-service approach manages the complexity of multi-device tracking automatically, allowing the system to handle device linking and data aggregation without requiring complex manual configuration or user input.
3Measurement precision
If device linking is performed using probabilistic matching, then accurate device relationships can be identified, but the time and computational resources required increase due to monitoring multiple linking factors
Solution Approach 1:
The system performs preliminary actions by pre-calculating and storing match probabilities for device relationships before content selection is needed. The probabilistic matching and device linking processes are executed in advance, creating a ready-to-use data structure that can be quickly applied during content selection. This preliminary action reduces the real-time processing burden, allowing the system to have device relationships pre-established rather than calculating them on-demand during content delivery.
Data Source
AI summary
A system includes a data processing system comprising one or more processors. The processor(s) are configured to identify one or more linking factors to determine that a first computing device is linked to a second computing device, with a certain probability, where the second computing device is proximate to the first computing device. The processor(s) are further configured to link the first computing device with the second computing device based on the determining, receive a request for content from the first computing device, provide the content to the second computing device, and generate a data structure to indicate a link between the first computing device and the second computing device.