Visual Schema Demand Prediction via Attribute Digitization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods fail to effectively quantify and analyze qualitative data such as product preference and demand using visual information from product images, as they require numerical representation of attribute relationships to predict consumer behavior.
Innovation Solution
A method is developed to create a visual schema by digitizing product attributes, analyzing these attributes using image analysis models, and expressing relationships through visual narrative data, enabling prediction and optimization of consumer demand and preference.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If visual information from product images is used to analyze consumer preference, then the ability to understand product attributes is improved, but the ability to quantitatively measure and predict demand deteriorates due to lack of numerical representation
Solution Approach 1:
The patent introduces an intermediary transformation layer that converts visual schema data into numerical representations. The visual schema, which captures semantic relationships between product attributes, serves as a mediator between raw image data and quantitative demand prediction models, enabling both visual understanding and numerical measurement.
Solution Approach 2:
The patent changes the parameter representation from qualitative visual attributes to quantitative numerical values. By transforming visual schema attributes into measurable parameters (e.g., attribute importance scores, relationship weights), the system enables statistical analysis and prediction while preserving the semantic meaning of visual information.
2Productivity
If attribute relationships are expressed numerically to enable prediction modeling, then demand prediction capability is improved, but the ability to capture complex visual relationships deteriorates
Solution Approach 1:
The visual schema acts as an intermediary that preserves visual relationship context while enabling numerical processing. It captures semantic connections between attributes (e.g., color-compatibility relationships, style associations) and represents them as structured data that can be quantified without losing interpretability.
Solution Approach 2:
The patent segments the complex visual attribute relationships into discrete, manageable components within the visual schema structure. By breaking down complex visual relationships into individual attribute connections with associated weights or probabilities, the system maintains detail while enabling numerical analysis.
3Measurement precision
If visual schema attributes are quantified to create numerical data for analysis, then the ability to perform predictive analytics is improved, but the complexity of data processing increases
Solution Approach 1:
The patent performs preliminary action by pre-processing and structuring visual data into a visual schema before it enters the predictive analysis pipeline. This preliminary organization of visual attributes into standardized relationships reduces the complexity of subsequent processing steps and enables more efficient numerical analysis.
Data Source
AI summary
The present disclosure relates to a method for predicting demand using a visual schema of a product, a device therefor, and a computer program therefor. The demand predicting method includes the operations of: creating visual schemas in which attributes of a product are digitized; analyzing the visual schemas and creating visual schema analysis data which are data relating to the attributes of the product; creating prediction data which are data obtained as a result of demand prediction analysis by attributes of the product using the visual schema analysis data; and creating visual narrative data expressing the prediction data into correlation between products or customers, and describing demand prediction.