Authors: Haihan Lin , Carolina Nobre , Amanda Bakian , Alexander Lex
Abstract: The analysis of multiple time series and associated quantitative or categorical attributes is an important task. Air pollution data, for example, is captured along many dimensions. This data is valuable for studying the association between air quality and diseases risks. In these cases, researchers need to view a large amount of data for multiple cases simultaneously, limiting the space that is available for each time series. In this abstract, we introduce clipped graphs, a hybrid clipped area chart that uses redundant color coding for visualizing time series data with skewed distributions and relatively rare peaks. We designed clipped graphs for use in compact tabular layouts. We use binned color scales for the full data range but clip outliers above a pre-defined threshold. The clipped peaks can be revealed through interaction. We integrate clipped graphs into an existing multivariate data visualization system, which visualizes clinical genealogies and detailed data about individuals in a tabular layout.