Abstract
We present a visualization tool called Structured Parallel Coordinates (SPC), a specialization of Parallel Coordinates (cf., e.g., Inselberg, 2009), customized for the presentation and analysis of different types of structured language data, as found in corpora. Parallel Coordinates are a way of representing multidimensional data using a two-dimensional display. Interactive versions of Parallel Coordinates are flexible tools for data analysis, since selecting parts of the visualization allows for filtering the data (Inselberg, 2009). Language datasets often have dimensions which are interrelated or which have internal structure, a situation that is not accounted for by standard Parallel Coordinates. We describe the visual features and interactions provided by SPC to account for structured language data and give technical details of the tool. We present three sample applications of SPC that are closely linked to principle tasks in corpus analysis: (1) KWIC results as SPC, (2) ngrams and frequencies, and (3) ranking comparisons.