Persistent Profile Identifiers for Stable Device Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional digital marketing tools are inadequate in identifying stable customer-device associations over extended periods, and existing clustering algorithms are impractical for large and complex datasets, leading to unstable and inefficient customer-device association tracking.
Innovation Solution
Assigning persistent profile identifiers to sets of devices by preserving maximum-matching identifiers across clustering runs, using a heuristic method for maximum cluster matching that reduces complexity and can process large datasets in a map-reduce context, ensuring stable customer-device associations for digital marketing strategies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional digital marketing tools are used to identify customer-device associations, then device identification is supported, but stable customer-device associations over extended periods cannot be achieved
Solution Approach 1:
The system performs preliminary clustering of devices into profiles before marketing campaigns, creating stable associations in advance. This allows the same customer profiles to be used across multiple clustering runs and extended time periods, rather than determining associations only during active campaigns.
Solution Approach 2:
The system creates copyable customer profile identifiers that can be reused across different clustering runs and time periods. These profile identifiers serve as stable representations of customer-device associations that can be replicated and maintained over extended durations.
2Measurement precision
If existing clustering algorithms are applied to large data systems, then customer-device associations can be identified, but computational complexity becomes impractically high
Solution Approach 1:
The system segments the large dataset into smaller clusters of devices that share common characteristics. By dividing the overall clustering problem into multiple smaller, manageable clusters, the computational complexity is reduced while maintaining association accuracy.
Solution Approach 2:
The system applies clustering iteratively to subsets of data rather than processing all data at once. Each clustering run processes a portion of the device population, and results are aggregated across multiple runs, making the overall process computationally feasible for large datasets.
3Quantity of substance
If existing clustering algorithms are used for large datasets, then associations can be determined, but scalability and efficiency are reduced
Solution Approach 1:
The system divides large datasets into smaller manageable clusters across multiple clustering runs. This segmentation enables parallel processing and reduces the computational burden on any single run, improving overall processing efficiency and scalability.
Solution Approach 2:
The system performs clustering periodically in discrete runs rather than continuously processing all data at once. Each periodic run processes a subset of data, and results are accumulated over time, enabling efficient handling of large datasets through batch processing.
Data Source
AI summary
Persistent profile identifiers can be produced to identify clusters of devices accessing a network in different time periods. In one embodiment, an apparatus uses a first identifier from a first group of identifiers to identify a first cluster of devices and uses a second identifier from a second group of identifiers to identify a second cluster of devices. Further, the apparatus determines that the first cluster of devices identified by the first identifier and the second cluster of devices identified by the second identifier form an edge in a maximum cluster matching. The apparatus provides the first identifier as a persistent identifier for the first cluster of devices and the second cluster of devices.