Binning Algorithm for Mixed Data Association Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to determine relationships and relevancy between large sets of complex, distributed analytics data that include both numeric and categorical data types, particularly in a distributed computing environment, where data is stored across multiple storage units and processed by multiple units.
Innovation Solution
The implementation of a binning algorithm to convert numeric data into categorical data, allowing for the comparison and analysis of both data types, followed by the computation of pairwise relationships and hierarchical clustering to identify strongly correlated data variables, which are then visualized through user interfaces such as heatmaps and chord diagrams.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data is stored across multiple storage units in a distributed computing environment, then data capacity and scalability are improved, but data complexity and difficulty of determining relationships increase
Solution Approach 1:
The patent segments the distributed data into categorical and numeric components, applying different processing strategies to each type. Categorical data is processed through frequency analysis and association rule mining, while numeric data undergoes statistical analysis and correlation computation. This segmentation allows the system to manage complex distributed data by breaking it into manageable, type-specific processing streams.
2Measurement precision
If diverse types of electronic data are collected from multiple sources, then user profile accuracy is improved, but data processing complexity increases
Solution Approach 1:
The patent transforms diverse data types into standardized parameters suitable for association analysis. Categorical data is converted into frequency distributions and contingency tables, while numeric data is transformed into statistical moments and correlation coefficients. This parameter transformation enables unified processing of heterogeneous data sources, improving user profile accuracy while managing processing complexity through standardization.
3Loss of information
If both categorical and numeric data are analyzed together, then comprehensive insights are improved, but computational difficulty increases
Solution Approach 1:
The patent introduces association rules and correlation coefficients as intermediary metrics that bridge categorical and numeric data analysis. Association rules capture relationships in categorical data through if-then statements, while correlation coefficients quantify relationships in numeric data. These intermediaries enable comprehensive analysis of mixed data types by providing common analytical frameworks that reduce computational difficulty compared to direct joint analysis.
Data Source
AI summary
Systems and methods disclosed herein identify multivariate relationships that exist across all types data collected from numerous observed users over one or more networks. Electronic data collected from observed users include categorical data and non-categorical/numeric data. To compare and analyze the collected data, a marketing entity converts the numeric data to categorical data via a binning algorithm, which reduces the numeric data into two or more discrete categories. The marketing entity analyzes the data variables to compute pairwise associations on the collected categorical and numeric data (which has been converted to categorical data). The marketing entity also determines hierarchical clusters to group the pairwise associations of data variables based on the strength of the associations. The pairwise relationships and hierarchical clusters are displayed on a user interface.