Rule Visualization System Using Multi-Plane Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing visualization techniques for large datasets of rules and exceptions are unwieldy and difficult to comprehend, especially when the number of antecedents and consequents increases, leading to occlusion and complexity in 3D representations, making it hard for end-users to understand and identify important rules.
Innovation Solution
A system and method for visualizing rules and exceptions using a processor and memory with modules for receiving, parsing, and visualizing data attributes, representing rules and exceptions with graphical elements that have proportional sizes and opacity based on support and confidence, and minimizing overlap by identifying cliques between nodes, allowing for semantic zooming and detailed exploration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If 3D landscape representation is used to visualize rules set, then more important rules can be placed in foreground and less important rules in background, but occlusion occurs hiding data points and the interface becomes complex and harder to understand
Solution Approach 1:
The visualization is segmented into multiple two-dimensional planes or layers, each representing different aspects of the ruleset (e.g., different consequents or rule categories). This segmentation eliminates occlusion by allowing users to navigate between planes rather than through a cluttered 3D space, while maintaining the ability to prioritize important rules within each plane.
Solution Approach 2:
Instead of using traditional 3D spatial representation that causes occlusion, the patent transitions to a multi-plane 2D representation system. This dimensionality change allows rules to be organized across multiple accessible planes with consistent scaling, eliminating perspective distortion and occlusion while preserving the ability to display large numbers of rules simultaneously.
2Quantity of substance
If the number of antecedents and consequents increases in the rules set, then more comprehensive coverage is achieved, but the visualization becomes unwieldy and difficult to understand
Solution Approach 1:
The large ruleset is segmented across multiple two-dimensional planes, with each plane focusing on a specific consequent or rule category. This allows comprehensive display of many rules while maintaining clarity within each plane, as users can navigate between planes to explore different aspects of the data without overwhelming visual complexity.
Solution Approach 2:
The visualization system implements dynamic plane selection and navigation, allowing users to interactively explore different planes based on their interests. Important rules can be dynamically highlighted or emphasized within planes, while less important rules remain visible but less prominent, adapting to user needs as they explore the comprehensive ruleset.
3Measurement precision
If perspective projection is used in 3D interface, then depth perception is provided, but sizes of objects are distorted
Solution Approach 1:
The patent eliminates perspective projection by transitioning from 3D to multi-plane 2D representation. This removes the source of size distortion while maintaining effective depth perception through plane layering and selective emphasis. Objects maintain consistent sizes regardless of their position, allowing accurate visual comparison of rule importance through controlled visual encoding rather than perspective-based size variation.
Data Source
AI summary
The present disclosure discloses system and method for providing perceptually efficient visualization of rules and exceptions mined from dataset. Further, parsing is performed on data-attributes associated with the rules. The data-attributes may include antecedents, consequents, ranges of the antecedents, syntax and statistics of the rules and exceptions. The visualization scheme of present disclosure present an overview first, allows semantic zooming, and then shows details on demand. Further, data attributes of the rules are mapped with visual attributes of graphical elements such as shape, color, opacity to create the perceptually efficient visualization of the rules and exceptions. Initially, the visualization shows main rule highlighting the exceptions associated and properties of the exceptions. Further, a semantic zoom slider is provided for allowing a user to navigate through different exception levels of the exception. Further, an interface is provided for obtaining additional information associated with the rules and the exceptions.