Title:Semantic Modeling of Analytic-based Relationships with Direct Qualification

Abstract:Successfully modeling state and analytics-based semantic relationships of documents enhances representation, importance, relevancy, provenience, and priority of the document. These attributes are the core elements that form the machine-based knowledge representation for documents. However, modeling document relationships that can change over time can be inelegant, limited, complex or overly burdensome for semantic technologies. In this paper, we present Direct Qualification (DQ), an approach for modeling any semantically referenced document, concept, or named graph with results from associated applied analytics. The proposed approach supplements the traditional subject-object relationships by providing a third leg to the relationship; the qualification of how and why the relationship exists. To illustrate, we show a prototype of an event-based system with a realistic use case for applying DQ to relevancy analytics of PageRank and Hyperlink-Induced Topic Search (HITS).