Invention Title:

System and Method to Generate and Enhance Dynamic Interactive Applications from Natural Language Using Artificial Intelligence

Publication number:

US20240386204

Publication date:
Section:

Physics

Class:

G06F40/284

Inventor:

Assignee:

Applicant:

Smart overview of the Invention

A novel system and method leverage artificial intelligence to generate and enhance dynamic interactive applications from natural language inputs. This approach involves capturing user requirements in a natural language format and utilizing AI to produce a functional application. The generated application is iteratively improved through ongoing feedback from users and the system, accommodating inputs in various formats such as natural, markup, programming, and domain-specific languages.

Background

Dynamic interactive applications are composed of dashboards featuring widgets, filters, and actions for optimal user experience. These dashboards interact with real-time data sources, allowing users to update information through action elements. Existing solutions like low-code platforms and AI-assisted tools partially automate application creation but often require expert intervention. Current systems struggle to fully automate the development of high-quality applications through natural language interactions.

Innovation

The disclosed system uses AI to transform natural language requirements into dynamic applications. It identifies necessary data sources and enriches requirements with domain-specific information. The enriched requirements undergo iterative summarization to fit AI processing capabilities, ensuring minimal information loss. A domain-specific language is employed to describe the application comprehensively, which the application generator uses to create a deployable product.

Deployment and Feedback

The AI-generated applications are deployed across various platforms, including web, desktop, VR/AR, and mobile. Automated and manual reviews ensure the application meets initial and evolving requirements. Feedback loops integrate user and system inputs in diverse formats to refine subsequent versions of the application. Version control is crucial for managing revisions and deploying updated applications effectively.

System Components

The system comprises several components: a requirements receiver for natural language input, an AI-based pre-processor for requirement analysis, a domain-specific enhancer for metadata enrichment, and a summarizer to optimize requirements for AI processing limits. The domain-specific language code generator creates applications based on summarized requirements, while an application builder converts these into functional applications. Feedback mechanisms ensure continuous enhancement by incorporating new requirements.