AI Source Scoring via Stretched Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to efficiently process queries for multiple items, leading to transport inefficiencies and potential item interoperability issues, as they fail to accurately prioritize sources based on item matching, quality, speed, and availability.
Innovation Solution
A computer-implemented method using artificial intelligence techniques, including natural language processing and machine learning, to determine a source metric for each item-source pair by transforming item-source metrics into stretch-normalized metrics, which are then used to generate source scores and prioritize sources based on requestor priorities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple items are processed individually from different sources, then item matching accuracy is improved, but transport efficiency deteriorates
Solution Approach 1:
The patent merges multiple individual item evaluations into a unified source scoring system. By combining item-source metrics for all items and applying stretch normalization to produce a single source score, the system achieves both accurate item matching and efficient transport planning in one integrated process.
Solution Approach 2:
The patent transforms raw item-source metrics into stretch-normalized metrics by applying a normalization factor that scales the metrics to a target range. This parameter transformation enables the system to handle multiple items of different types and priorities uniformly, resolving the contradiction between precise matching and efficient processing.
2Measurement precision
If sources are ranked based on individual item metrics, then source selection accuracy is improved, but processing complexity increases
Solution Approach 1:
The patent segments the complex evaluation process into distinct steps: calculating item-source metrics for each item-source pair, aggregating these metrics into source scores, and applying stretch normalization. This segmentation reduces processing complexity by breaking down the overall task into manageable, sequential operations.
Solution Approach 2:
The patent creates a universal source scoring function that can evaluate multiple items of different types and priorities using the same stretch normalization process. This multi-functional approach simplifies processing by using a single standardized method rather than separate evaluation procedures for each item type.
3Measurement precision
If stretch normalization is applied to item-source metrics, then source score comparability is improved, but computational operations increase
Solution Approach 1:
The patent performs stretch normalization as a preliminary action during the source scoring process. By normalizing the metrics early in the computation pipeline, the system ensures comparable source scores are ready for subsequent processing steps without adding computational complexity later in the workflow.
Data Source
AI summary
The present disclosure relates to systems and methods for using an artificial intelligence technique for determining a source score based on stretched normalization. A natural language query can be received and mapped. Sources can be identified, and actions can be taken with respect to each source. The actions can include determining an item-source metric, transforming the item-source metric using a stretched-normalization factor, and generating a source score based on the transformed item-source metric. A response to the natural language query can be generated based on the source score, and the response can be output.