Natural Language AI Agent for Prescriptive Model API Execution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
End users lack the skills to operate prescriptive AI models, and data science groups and business-focused users often work in silos, impeding swift execution, while prescriptive AI models are typically domain-specific and trained with application-dependent data absent from standard training data of other AI models.
Innovation Solution
A computer-implemented method that receives natural language inputs, matches them to API calls for prescriptive tasks, identifies parameter values, and executes prescriptive AI models to generate responses, utilizing large language models for intent classification and slot filling to ensure necessary data is obtained.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If prescriptive AI models are made domain-specific with application-dependent data, then model accuracy and relevance improve, but accessibility and ease of operation deteriorate because end users lack the skills to operate these models
Solution Approach 1:
The patent introduces an intermediary system that includes a natural language processing interface and an API call generation mechanism. This intermediary translates user-friendly natural language inputs into the complex parameter structures required by domain-specific prescriptive AI models, thereby maintaining high model accuracy while improving user accessibility without requiring users to have specialized skills
Solution Approach 2:
The system segments the complex interaction into two distinct layers: a user-facing natural language interface layer and a backend model execution layer. This segmentation allows the prescriptive AI model to remain domain-specific and accurate while the interface layer handles the complexity of data preparation and parameter mapping, making the system accessible to end users
2Reliability
If data science groups and business-focused users work in silos, then each group maintains specialized expertise, but execution speed and collaboration deteriorate
Solution Approach 1:
The patent creates a universal interface system that serves multiple functions: it accepts natural language from business users, automatically generates appropriate API calls, and interfaces with prescriptive AI models. This multi-functional system eliminates the need for direct collaboration between data science groups and business users, allowing both to maintain their specialized roles while achieving swift execution through the universal interface
3Loss of time
If standard training data of AI models is used, then model development time is reduced, but domain-specific accuracy and relevance deteriorate
Solution Approach 1:
The system performs preliminary action by pre-configuring API calls with the correct structure, parameter mappings, and data formats required by domain-specific prescriptive AI models. This preliminary setup allows the system to quickly process user inputs without requiring time-consuming model retraining, thereby reducing development time while maintaining domain-specific accuracy through pre-established connections to specialized models
Data Source
AI summary
An example operation may include one or more of receiving at least one natural language input via a software application, determining that the at least one natural language input matches an application programming interface (API) call from among a plurality of API calls configured for prescriptive tasks, identifying at least one parameter value of the API call from the at least one natural language input and transmitting the API call to a prescriptive artificial intelligence (AI) model, executing the prescriptive AI model on the at least one parameter value to generate a natural language response, and displaying the natural language response via a graphical user interface (GUI) of the software application.