Relaxed Bi-Clustering for User Segment Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional analytics systems face challenges in accurately and efficiently identifying user segments in high-dimensional data spaces due to noise and inflexibility, often requiring significant processing power and manual coding, which can lead to inaccurate and uninterpretable results.
Innovation Solution
A digital analytics system employs a relaxed bi-clustering model that merges and filters user segments to create meaningful and insightful reports, allowing users to be included even if they don't share all features, thereby overcoming the limitations of rigid clustering techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional analytics systems use rigid clustering techniques to identify user segments, then processing power consumption increases, but the accuracy and flexibility of segment identification deteriorates
Solution Approach 1:
The patent changes the parameters of clustering from rigid requirements (all binary entries must be one) to relaxed requirements (allowing some entries to be zero). This parameter change enables the system to identify user segments with fewer strict constraints, reducing processing power consumption while maintaining or improving accuracy by including adjacent users who share many characteristics.
Solution Approach 2:
The patent applies partial action by allowing clustering to proceed with incomplete or approximate matches rather than requiring complete matches. The relaxed bi-clustering model identifies segments where users share a substantial portion of characteristics rather than all characteristics, reducing the computational burden while still producing meaningful segments.
2Productivity
If conventional analytics systems manually program code to generate segments for high-volume websites, then processing power requirements are met, but time consumption increases and flexibility to revise segments is reduced
Solution Approach 1:
The patent implements self-service through automatic segment generation using the relaxed bi-clustering model. The system autonomously identifies user segments without requiring manual programming of code, significantly reducing the time and expertise needed while maintaining high processing capability for analyzing large volumes of data.
Solution Approach 2:
The patent introduces dynamics by enabling easy revision and reconfiguration of segments. The relaxed clustering model allows administrators to modify segment criteria and re-run analysis quickly without complex reprogramming, making the system adaptable to changing business needs while maintaining high productivity.
3Adaptability or versatility
If conventional analytics systems use rigid clustering where every binary entry must be one, then processing is simplified, but the ability to include adjacent users with similar characteristics is lost
Solution Approach 1:
The patent changes the parameter requirements from rigid (all entries must be one) to relaxed (allowing some entries to be zero). This enables the inclusion of adjacent users who share many characteristics but not all, increasing adaptability while the algorithmic complexity remains manageable through the bi-clustering approach.
Data Source
AI summary
Systems, methods, and non-transitory computer-readable media (systems) are disclosed for generating meaningful and insightful user segment reports based on a high dimensional data space. In particular, in one or more embodiments, the disclosed systems utilize a relaxed bi-clustering model to automatically identify user segments in a data space including datasets of features specific to individual users. In at least one embodiment, the disclosed systems identify and include users in automatically generated user segments even though those users are associated with some, but perhaps not all, of the features as other members in the automatically generated user segments.