Enhancing Visualization Applications Using Open Data Sources

An increasing number of data visualization tools have started to support the automatic generation of modifications, embellishments, and natural language annotations (data facts) to aid in better understanding the data being visualized. Concurrently, many applications in data science now benefit from the use of an increasingly diverse set of open data sources to augment existing data sets to enhance their value. In this paper we present a framework for using open data-based augmentations to generate embellishments and data facts to enhance existing visualizations. Our approach is based on a semi-automated process intended to involve the user, where possible augmentations are automatically ranked based on the data facts they are capable of generating, allowing users to choose augmentations to effectively enhance existing data visualizations in an explorative manner. We show the benefit of suggesting ranked augmentations from one open data source, Wikidata, by demonstrating that a high number of data facts and embellishments can be produced utilizing the top suggested augmentations. Finally, we describe the architecture of a prototype system implementing the approach.