Cross-Device User Category Synchronization via Device Graph
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for cross-device user category synchronization in online advertising campaigns are inefficient, particularly in real-time scenarios, as they struggle to associate users across multiple devices without relying on browser cookies, leading to missed opportunities in targeted advertising.
Innovation Solution
The implementation of systems and methods for real-time cross-device user category synchronization, which involve identifying user data records from various devices, associating them based on matched attributes, and forming a cross-device map to synchronize user categories, thereby enabling targeted advertising across all devices used by a user.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If legacy cross-device matching methods are used to associate users across multiple devices, then user profile information can be captured, but real-time synchronization of user categories is inefficient and requires excessive memory usage, processing power, and communication overhead
Solution Approach 1:
The patent segments the cross-device matching process into distinct phases: device graph construction phase (offline) and synchronization phase (online). The device graph is pre-built and stored in a data structure optimized for rapid querying, separating the computationally intensive matching operations from the real-time synchronization operations. This allows real-time synchronization to simply query pre-computed relationships rather than performing full matching algorithms.
Solution Approach 2:
The patent performs preliminary cross-device matching to construct a device graph before real-time synchronization is needed. User categories and device relationships are pre-computed and stored in an optimized data structure, so that during real-time synchronization, the system only needs to perform efficient lookups rather than complex computations. This preliminary action eliminates the need for excessive processing power during real-time operations.
2Measurement precision
If comprehensive device attributes are collected and processed for cross-device matching, then user association accuracy improves, but communication overhead and data processing requirements increase
Solution Approach 1:
The patent extracts only the essential device attributes needed for matching into a compact device graph data structure, separating critical matching features from comprehensive device profiles. The device graph stores only the attributes necessary for matching (such as device identifiers, hashed attributes, and relationship pointers) rather than complete device information, reducing communication overhead while maintaining matching accuracy.
Solution Approach 2:
The patent creates a simplified copy of device information in the form of a device graph, which is a lightweight data structure containing only the essential matching attributes and relationships. This copy allows rapid comparison and matching operations without requiring access to or transmission of the full device profiles, thereby reducing communication overhead while preserving user association accuracy.
3Reliability
If real-time cross-device user category synchronization is implemented, then targeted advertising effectiveness improves, but processing time and computational resources increase
Solution Approach 1:
The patent implements a dynamic system where the device graph is constructed offline and then used for rapid real-time synchronization. The system adapts between offline batch processing (for graph construction) and online real-time processing (for synchronization), allowing it to maintain high advertising effectiveness while minimizing real-time processing time. The pre-built device graph enables the system to respond dynamically to real-time synchronization requests with minimal computational delay.
Data Source
AI summary
The present disclosure provides systems, methods, and computer program products for real-time cross-device user category synchronization. Embodiments commence upon allocating or identifying user data records comprising two user profiles. The user profiles are populated with information by receiving electronic communications over a network, where the information comprises user category information and device attributes found in data records that originate from user devices. Steps are taken to associate a first device data record with a second device data record, wherein the association is based at least in part on a match between selected attributes among the user devices. User category information within the user profiles are synchronized to form updated user profiles. Some embodiments further execute steps for forming a cross-device map that identifies an association between an updated first user profile and a second user profile.