Multi-Dimensional Product Information Visualization via Sorting Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods fail to effectively visualize multi-dimensional product information from different categories, limiting users' ability to intuitively grasp and differentiate attribute information across various products.
Innovation Solution
A product information visualization processing method and apparatus that acquires and extracts attribute data sets, inputting them into a pre-trained sorting model to identify and output sorting results, utilizing an encoder-decoder architecture with attention mechanisms and reinforcement learning for efficient visualization of high-dimensional data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional product information processing methods are used, then the processing is simple and straightforward, but the visualization of multi-dimensional data from different product categories is ineffective and lacks intuitive global perspective
Solution Approach 1:
The patent segments the complex multi-dimensional product information into distinct attribute dimensions (e.g., price, performance, features). Each dimension is processed and visualized separately through dedicated sorting models, allowing precise visualization of each attribute while managing complexity through modular processing units.
Solution Approach 2:
The patent transforms multi-dimensional product data into a visual space by mapping attributes to different visual dimensions (positions, sizes, colors, shapes). This dimensionality transformation enables intuitive global comparison across product categories by representing abstract attributes in spatial relationships that human cognition can easily process.
2Loss of information
If multi-dimensional attribute data of different product categories is processed, then comprehensive product information is obtained, but effective visualization difference evaluation cannot be performed
Solution Approach 1:
The patent applies different visualization strategies and sorting models to different attribute dimensions based on their local characteristics. For example, numerical attributes like price use sorted position encoding, while categorical attributes use color or shape encoding. This local optimization ensures that each attribute type is visualized in the most effective way while maintaining overall information completeness.
Solution Approach 2:
The patent dynamically changes visualization parameters (position, size, color, shape) based on the attribute type and data characteristics. Sorting models adjust these parameters to emphasize differences between product categories, making it easier for users to evaluate and compare products across multiple dimensions without information loss.
3Ease of operation
If human cognitive ability is considered limited, then the processing should be simplified, but the current methods cannot provide intuitive global perspective on multi-dimensional product information
Solution Approach 1:
The patent uses color encoding to represent different attribute dimensions and product categories. By assigning distinct colors to different attributes (e.g., price, performance, features), the system enables users to quickly distinguish and compare multiple dimensions simultaneously, reducing cognitive load while maintaining precise information discrimination through color-coded visual cues.
Solution Approach 2:
The patent maps multi-dimensional product attributes to spatial dimensions in the visualization (position, size, shape). This transformation leverages human spatial cognition strengths, allowing users to intuitively grasp global relationships and differences across product categories by visually comparing positions and sizes rather than processing abstract numerical data.
Data Source
AI summary
A product information visualization processing method, including: acquiring product information; extracting an attribute data set of multiple dimensions corresponding to the product information; and inputting the attribute data set of the multiple dimensions into a pre-trained sorting model, and identifying the attribute data set of each dimension by using the sorting model until sorting results corresponding to the multiple dimensions are outputted according to a preset number of attribute dimensions.