Enhancing Information Awareness Through Directed Qualification of Semantic Relevancy Scoring Operations

TL;DR

This paper introduces a method to improve semantic relevancy scoring by integrating analytics-based relationship qualification using ontologies, enabling more precise document importance assessment.

Contribution

It presents a novel approach combining Prov-O and relevancy ontologies to model qualified, analytics-derived relationships in semantic technologies.

Findings

Enables association of documents with analytics results as qualified relationships.

Supports role, identity, and content-based relevancy metrics.

Enhances semantic models with autonomous qualification capabilities.

Abstract

Successfully managing analytics-based semantic relationships and their provenance enables determinations of document importance and priority, furthering capabilities for machine-based relevancy scoring operations. Semantic technologies are well suited for modeling explicit and fully qualified relationships but struggle with modeling relationships that are qualified in nature, or resultant from applied analytics. Our work seeks to implement the autonomous Directed Qualification of analytic-based relationships by pairing the Prov-O Ontology (W3C Recommendation) with a relevancy ontology supporting analytics terminology. This work results in the capability for any semantically referenced document, concept, or named graph to be associated with the results of applied analytics as Direct Qualification (DQ) modeled relational nodes. This new capability will enable role, identity, or any other content-based measures of relevancy and analytics-based metrics for semantically described documents.