Hierarchical Tree Data Structure for Multi-Dimensional Visualization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data structures, such as traditional join model databases, face limitations in efficiently managing and visualizing data with multiple dimensions and measures, particularly in generating hierarchical charts and handling complex data representations.
Innovation Solution
The implementation of a hierarchical tree data structure that stores measures as values within nodes, allowing data to be calculated based on current and parent levels, and enabling the storage of multiple measures in any level, which simplifies data representation and visualization by merging levels and eliminating the distinction between different types of levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional join model databases are used to manage multi-dimensional data, then data can be stored and retrieved, but the system becomes complex and inefficient when generating hierarchical charts and handling complex data representations
Solution Approach 1:
The patent merges multiple levels of data hierarchy into a single flattened level, combining what were previously separate hierarchical levels into one unified structure. This eliminates the need for complex multi-level navigation and joins, simplifying both data storage and visualization generation while maintaining all dimensional relationships.
Solution Approach 2:
The patent transforms the traditional multi-level hierarchical structure into a single-level structure by adding dimensional attributes to each node. Instead of organizing data vertically across multiple levels, the invention reorganizes it horizontally with all dimensions present in one level, changing the structural dimensionality from vertical hierarchy to horizontal attribution.
2Reliability
If multiple levels are maintained in hierarchical data structures to preserve data relationships, then data integrity is maintained, but the number of nodes and levels increases, reducing efficiency
Solution Approach 1:
The patent combines multiple hierarchical levels into a single level while preserving all parent-child and sibling relationships through dimensional attributes. Each node contains references to its dimensions and relationships, maintaining data integrity without requiring physical multi-level structure, thus improving processing efficiency.
Solution Approach 2:
The patent creates a universal single-level node structure that can represent any hierarchical relationship through its dimensional attributes. Each node is multi-functional, capable of representing parent, child, or sibling relationships depending on its dimensional configuration, eliminating the need for specialized structures for different relationship types.
3Productivity
If data is stored in a flattened single level structure, then processing efficiency improves, but data set size may increase
Solution Approach 1:
The patent merges hierarchical levels by storing all dimensional information within single-level nodes rather than distributing it across multiple levels. This consolidation eliminates redundant data that would exist in traditional hierarchical representations, as each node independently contains all its dimensional attributes and relationships without requiring duplicate structures in parent or child levels.
Data Source
AI summary
Methods and systems are disclosed for generating and using a hierarchical tree data structure. A computing device may receive data having a number of dimensions and corresponding values for each of the dimensions. A hierarchical tree data structure may have a root node and a number of levels, one of which may be connected to the root node. Each level may correspond to a respective dimension of the data and may be connected to a higher level as well as to a lower level. A measure for a given dimension may be evaluated based on the nodes corresponding to the level that corresponds to the given dimension and all preceding connected nodes. A result of the measure may be stored in a number of nodes corresponding to the level that corresponds to the given dimension.