ABSTRACT
How can this design space be used? For structured data, text can add data to points in common charts such as bar charts, scatterplots, Venn diagrams, and maps; or generate text as sentences and paragraphs enriched with structured data embedded into formats. For unstructured data, such as raw text from prose, natural language processing can be used to extract data. In turn, this data can be overlaid onto the prose; or words and phrases from the prose can be assembled into visualization such as graphs, lists, or tables. When embedding data into text formats, the same data may be redundantly encoded to multiple attributes, such as size and hue, to aid differentiation and improve legibility; or multiple attributes may encode different data to aid visual inspection across a conjunction of features. These textual formats may extend to text outside the plot area, such as axes and tick labels, annotations and data comics, data narratives, and lists. Design spaces can also be used to find potential new visualization applications, such as monitoring, search, NLP, markup, augmented reality, and knowledge mapping.
