AI-Guided Data Resource Mapping for Reliable Metric Calculation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In large or complex organizations, manually tracking and identifying relevant data resources for business analytics tasks is time-consuming due to the numerous non-standard data resources, and existing AI techniques require manual data identification for accurate operation.

Innovation Solution

A computer-implemented method and system that uses a user-guided approach with AI to automatically identify and validate data resources and metrics through natural language queries, generating an instruction set for a deterministic function to compute metrics from data resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual tracking and identification of data resources is used, then users can accurately locate relevant data, but the process becomes time-consuming and inefficient due to the large number of nonstandard data resources

Engineering Contradiction:

Improveaccuracy of data resource identificationVSAvoidtime required to locate relevant data

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical searching and tracking of data resources with an AI-based automated system. The AI model generates embeddings for both data resources and user queries, enabling automatic matching and identification of relevant data resources without manual intervention, thus resolving the contradiction between identification accuracy and time consumption

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent transforms data resources and queries into a common parameter space using embeddings. By converting both data resources and user queries into vector representations, the system enables efficient similarity-based matching, allowing rapid identification of relevant data resources while maintaining accuracy through mathematical distance metrics in the embedding space

Inventive Principle:

Principle #35Parameter changes

2Productivity

If AI techniques are used to automate analytics operations, then processing speed and efficiency increase, but reliable operation still requires manual identification of correct data resources

Engineering Contradiction:

Improvespeed of analytics processingVSAvoidease of data resource identification

Core Design Contradiction:

ProductivityVSEase of operation

Solution Approach 1:

The patent implements a self-service system where the AI model autonomously performs data resource identification, validation, and metric computation. The system automatically generates embeddings, matches queries with data resources, validates the mappings, and computes metrics without requiring manual data identification, thereby maintaining both high productivity and ease of operation

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent introduces embeddings as an intermediary layer between user queries and data resources. This embedding layer acts as a mediator that translates both queries and data resources into a common representation space, enabling the AI system to automatically bridge the gap between user intent and relevant data without manual intervention

Inventive Principle:

Principle #24Intermediary (Mediator)

3Reliability

If users have extensive knowledge of data collection to select relevant resources, then accurate metrics can be derived, but the system becomes complex and difficult to operate for users without specialized knowledge

Engineering Contradiction:

Improvereliability of metric computationVSAvoidcomplexity of data resource selection process

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent replaces the need for user expertise in data resource selection with an AI-based automated selection mechanism. The system automatically generates embeddings for data resources and queries, performs similarity matching, and validates mappings, thereby ensuring reliable metric computation without requiring users to understand the underlying data collection complexity

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The embedding-based matching system serves as an intermediary that handles the complexity of data resource selection. By translating both data resources and queries into embeddings and performing automatic similarity matching, the system shields users from the underlying complexity while maintaining reliable metric derivation through accurate data-resource pairing

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS20260064736A1Data resource identification and metric calculation

Publication Date: 2026.03.05 SAGE GLOBAL SERVICES LTD

AI summary

A computer implemented method for deriving a metric from a data resource. A data resource query is received from a user device and mapped to one or more data resources. Identifiers associated with mapped data resources are communicated to the user device. A mapping validation is received from the user device, representing a user validation of the mapping of data resources to the data resource query. A metric query is received from the user device and is similarly mapped to metrics with the mapping being validated by the user device. The validated mapped data resource and data defining how to compute the validated mapped metric are retrieved. A composite query including the metric definition data and the retrieved data resource and an instruction for an AI model to generate an instruction set for a deterministic function that applies the metric computation to the data resource are retrieved.