Data‐driven semantic indexing of digital assets documenting American football game action

Abstract

AbstractThis is a work‐in‐progress poster reporting on research investigating the item‐level semantic indexing of digital assets (i.e., images and video clips) that document action during American football games. Our data‐driven semantic indexing method is novel in that it is not based on the traditional asset‐by‐asset approach that results in the creation of individual metadata records for each asset; rather, our method inverts the traditional process by facilitating the organizing of each game on a play‐by‐play basis first by using play‐level statistical datasets for those games, and then “attaching” each digital asset to the appropriate play via an actual or simulated time‐based parameter. In this poster, we will report on the implementation of our MediaWiki/Wikibase instance, on key classes and properties from our data model, and we will provide a live demo of example SPARQL queries demonstrating the precision searches that are made possible by our data‐driven semantic indexing method.