Chart De-Rendering AI Evaluation Using Synthetic Ground Truth Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for constructing data sets for chart de-rendering AI models suffer from limited data diversity, lack of meta information, and difficulty in categorizing charts by style, which hampers accurate evaluation and training of the models.
Innovation Solution
A system and method that generates a large amount of data sets including chart information, incorporating line and meta information, allowing for diverse chart styles and accurate classification into experimental and control groups, using a data set generation model to store GT and output data formats for performance evaluation and training.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data is collected by crawling a specific website, then data can be obtained, but the amount of data is limited and diversity is lacking
Solution Approach 1:
The patent uses a data set generation model to synthetically generate chart images and their corresponding ground truth data, copying the structure and characteristics of real chart data without requiring actual crawling. This approach produces unlimited diverse data while maintaining the statistical properties needed for training, directly resolving the contradiction between data quantity and diversity.
2Ease of manufacture
If conventional crawling methods are used, then data collection is simple, but meta information is not properly included
Solution Approach 1:
The data set generation model pre-generates complete ground truth data including all necessary meta information (chart titles, axis labels, legends, data values) alongside the synthetic chart images. This preliminary inclusion of complete metadata in the training data structure ensures the AI model receives full information without requiring complex post-processing or information recovery.
3Measurement precision
If chart images of similar style are classified into experimental and control groups, then accurate evaluation is possible, but difficulty in collecting sufficient similar-style charts arises
Solution Approach 1:
The data set generation model uses parameter control to generate chart images with consistent styles while varying other attributes. By fixing style parameters (e.g., chart type, color scheme, layout) and varying data parameters, the system produces large numbers of similar-style charts suitable for controlled experimental and control group comparisons, enabling accurate evaluation without manual collection constraints.
Data Source
AI summary
A system, method, and program evaluate the performance of an artificial intelligence (AI) model that de-renders a chart or for training the AI model by constructing a data set including chart information. The system includes memory storing a data set generation model and an AI model, and a processor configured to execute or train the AI model and execute a performance evaluation model. The data set generation model stores line information, which is information about a line of a chart, and meta information, which is information about meta data, as ground truth (GT), stores an image formed using the GT as a chart image, and outputs the GT and the chart image as a data set, and the AI model receives the chart image stored in the data set as input and outputs a data format in which information of the chart is predicted.