ML Database Schema Generation from UI Mockups

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current software development processes require significant manual effort and are prone to errors when designing and implementing database schemas that reflect the user interface (UI) design of applications.

Innovation Solution

A machine learning-based system that uses computer vision and natural language processing techniques to automatically generate a structured database schema from UI mockups and associated textual annotations and labels.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If manual effort is used to design database schemas, then flexibility and adaptability are maintained, but time consumption and error rates increase

Engineering Contradiction:
Improvedatabase design speedVSAvoidschema accuracy
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent replaces manual mechanical database schema design with an automated machine learning system that uses computer vision and natural language processing. The system automatically extracts information from UI mockups and generates database schemas, eliminating manual typing and reducing human error while maintaining accuracy through intelligent pattern recognition and data mapping algorithms.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service database schema generation where the machine learning model automatically processes UI mockups, extracts relevant information, and creates database schemas without requiring manual intervention. The system serves itself by autonomously completing the database design task that would otherwise require skilled manual effort.

Inventive Principle:
Principle #25Self-service

2Productivity

If automated systems are used to generate database schemas, then productivity increases, but complexity of the system increases

Engineering Contradiction:
Improvedatabase design speedVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the complex automated system into distinct functional modules: a computer vision component for processing UI mockup images, a natural language processing component for extracting textual information, and a database schema generation component for creating the final output. This segmentation makes the overall complex system more manageable and easier to implement by breaking it down into smaller, well-defined tasks.

Inventive Principle:
Principle #1Segmentation

3Loss of time

If manual database schema design is used, then ease of operation is maintained, but loss of time increases

Engineering Contradiction:
Improvedatabase design timeVSAvoidoperational simplicity
Core Design Contradiction:
Loss of timeVSEase of operation

Solution Approach 1:

The patent introduces machine learning models as intermediaries between the UI mockup design and the database schema. The computer vision model and natural language processing model act as mediators that automatically translate visual and textual information from the mockup into structured database schema, eliminating the need for manual translation and significantly reducing time loss.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS12298947B2Machine-learning based generation of database configurations for databases associated with applications
Publication Date: 2025.05.13 DELL PROD LP
  • US12298947B2 patent drawing
  • US12298947B2 patent drawing
  • US12298947B2 patent drawing

AI summary

An apparatus comprises at least one processing device configured to obtain information characterizing a user interface (UI) of an application, the obtained information comprising an image of the UI and textual data associated with the UI. The processing device is also configured to generate, by processing the image utilizing a first machine learning model, a first data structure characterizing data fields corresponding to UI components in the UI. The processing device is further configured to generate, by processing the textual data utilizing a second machine learning model, a second data structure characterizing relationships between the data fields. The processing device is further configured to determine, based on the first and second data structures, a database configuration comprising an organization of the data fields into tables for a database associated with the application. The processing device is further configured to generate the database in accordance with the determined database configuration.