AI/ML Visualization Generation Through Deterministic Data Retrieval

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing AI/ML models struggle with generating consistent and accurate visualizations, often producing inconsistent styles and inaccurate representations of data, particularly when dealing with large datasets, and lack the precision needed for reliable data visualization.

Innovation Solution

A system that leverages AI/ML models to generate data retrieval instructions, such as SQL statements, which are then translated into visualization properties by a non-AI/ML module, ensuring consistent styles and accurate data representation by combining the strengths of AI/ML models with non-probabilistic data processing systems like database management systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If AI/ML models are used to generate visualizations directly, then the system can process natural language input and generate visualization outputs, but the visualizations become inconsistent in style and inaccurate in data representation

Engineering Contradiction:

Improvenatural language processing capabilityVSAvoidvisualization accuracy and consistency

Core Design Contradiction:

Ease of operationVSManufacturing precision

Solution Approach 1:

The system divides the visualization generation process into distinct segments: (1) AI/ML model generates data retrieval code from natural language, (2) Code executor retrieves accurate data from database, (3) Visualization generator creates consistent visualizations from structured data. This segmentation allows each component to optimize for its specific function, with the AI handling natural language and the data processing system ensuring accuracy.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary layer consisting of code execution and data retrieval components between the AI/ML model and the final visualization. This intermediary ensures that the probabilistic output of the AI model is transformed into deterministic, accurate data queries that guarantee consistent and precise visualizations.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Adaptability or versatility

If AI/ML models generate visualization code, then the system can leverage generative capabilities, but the code may not reliably represent values and proportions accurately

Engineering Contradiction:

Improvegenerative capabilityVSAvoiddata representation accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent replaces the mechanical generation of visualization code with a substitution approach where the AI model generates data retrieval code (SQL queries), which is then executed by a database system to obtain accurate data. This substitution ensures that the measurement precision requirements are met by leveraging the database system's exact data retrieval capabilities rather than relying on the AI model's probabilistic generation.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Reliability

If the system uses multiple interactions with AI/ML models, then the system can combine repeatable data retrieval with generative capabilities, but the process complexity increases

Engineering Contradiction:

Improverepeatable accurate data retrievalVSAvoidsystem process complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent creates a universal framework where the same architecture can handle various visualization types and data sources. The AI model serves multiple functions (natural language understanding, code generation, data querying), and the system can consistently produce accurate visualizations across different scenarios without requiring separate specialized components for each function.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS20250336129A1Generating visualizations using artificial intelligence or machine learning

Publication Date: 2025.10.30 STRATEGY INC

AI summary

Methods, systems, and apparatus, including computer programs encoded on computer-storage media, for managing artificial intelligence chatbots. In some implementations, a system identifies a data source with which to generate a visualization. The system generates a request for an artificial intelligence or machine learning (AI/ML) model to generate code or instructions to retrieve from the data source data that satisfies one or more criteria. The system sends the request and receives code or instructions that the AI/ML model generated in response to the request. The system determines one or more visualization properties based on the code or instructions that the AI/ML model generated in response to the request. The system provides visualization data for a visualization of data from the data source presented according to the one or more visualization properties.