Post-Modeling Data Visualization with Key Feature Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data visualization methods do not provide adequate insights into specific portions of data selected by users, failing to offer evaluations or detailed analyses of user-selected data groups.
Innovation Solution
A method and system for post-modeling data visualization and analysis that identifies three or fewer key model features from a user-selected data group, using techniques like Local Interpretable Model-Agnostic Explanation (LIME) to ascertain representative records, and presents these in a second visualization plot for deeper analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If data visualization presents comprehensive data and relationships, then data insights are improved, but visualization complexity increases
Solution Approach 1:
The patent segments the data visualization into two distinct plots: a first plot displaying comprehensive training dataset relationships, and a second plot displaying selected data group details with key model features. This segmentation allows users to navigate from overall data patterns to specific insights without overwhelming complexity in a single visualization.
Solution Approach 2:
The patent applies local quality by providing detailed analysis of specific data groups (local) while maintaining overview of the entire training dataset (global). The second plot focuses on key model features and representative records for selected data groups, providing localized depth without sacrificing overall context.
2Loss of information
If the system analyzes all data groups, then analysis completeness is improved, but processing time increases
Solution Approach 1:
The system performs preliminary actions by pre-identifying key model features and pre-processing data groups before user interaction. The first visualization establishes baseline understanding of data relationships, and when users select specific data groups, the system has already prepared the framework for detailed analysis, reducing real-time processing requirements.
Solution Approach 2:
The patent applies partial action by analyzing only selected data groups rather than all data groups simultaneously. The system identifies three or fewer key model features for the selected data group, providing sufficient analysis depth without the computational burden of processing every data group in the entire training set.
3Measurement precision
If the system identifies multiple key model features, then feature representation accuracy is improved, but visualization clarity deteriorates
Solution Approach 1:
The patent applies local quality by providing detailed analysis of specific data groups (local) while maintaining overview of the entire training dataset (global). The second plot focuses on key model features and representative records for selected data groups, providing localized depth without sacrificing overall context.
Data Source
AI summary
In an approach for post-modeling data visualization and analysis, a processor presents a first visualization of a training dataset in a first plot. Responsive to receiving a selection of a data group of the training dataset to analyze, a processor identifies three or fewer key model features of the data group of the training dataset. A processor ascertains a representative record of each key model feature of the three or fewer key model features using a Local Interpretable Model-Agnostic Explanation technique. A processor presents a second visualization of the three or fewer key model features and the representative record of each key model feature in a second plot.