Content Entity Annotation Scoring with Linear Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing content annotation systems fail to efficiently satisfy joint quality, coverage, and completeness criteria, leading to suboptimal entity annotation for content processing tasks such as search engine results and targeted advertisements.
Innovation Solution
A system and method that includes a processor and memory with computer executable components for aggregating and mapping initial entity scores, applying weights based on joint performance conditions, and combining scores using a linear aggregation model to determine whether to annotate content with entities, ensuring quality, coverage, and completeness criteria are met.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional content annotation systems are used, then annotation processing can be performed, but the systems fail to efficiently satisfy joint quality, coverage, and completeness criteria
Solution Approach 1:
The patent transforms the annotation scoring process by changing the parameter representation method. It maps initial scores from multiple annotation sources to a unified calibrated score scale, then applies linear aggregation with optimized weights to combine scores. This parameter transformation enables simultaneous optimization of quality, coverage, and completeness metrics while maintaining processing efficiency.
2Quantity of substance
If multiple content annotation sources are aggregated, then more entities can be identified, but the complexity of satisfying joint performance conditions increases
Solution Approach 1:
The patent introduces an intermediary calibration layer between multiple annotation sources and the final aggregation step. The calibration component transforms diverse score distributions from different sources into a unified scale, serving as a mediator that simplifies the subsequent linear aggregation process. This intermediary structure enables handling multiple sources without proportionally increasing system complexity.
3Productivity
If linear aggregation models are used to combine scores, then computation remains efficient, but the ability to capture complex relationships between annotation sources is limited
Solution Approach 1:
The patent performs preliminary calibration of scores from different annotation sources before applying linear aggregation. By pre-processing the scores to establish a unified scale and pre-computing optimal weights based on source reliability, the system prepares the data in advance. This preliminary action enables the use of simple linear aggregation while achieving high precision, as the complex relationships are resolved in the calibration and weight optimization stages.
Data Source
AI summary
Facilitating of content entity annotation while maintaining joint quality, coverage and/or completeness performance conditions is provided. In one example, a non-transitory computer-readable medium comprises computer-readable instructions that, in response to execution, cause a computing system to perform operations. The operations include aggregating information indicative of initial entities for content and initial scores associated with the initial entities received from one or more content annotation sources and mapping the initial scores to respective values to generate calibrated scores. The operations include applying weights to the calibrated scores to generate weighted scores and combining the weighted scores using a linear aggregation model to generate a final score. The operations include determining whether to annotate the content with at least one of the initial entities based on a comparison of the final score and a defined threshold value.