AutoVis: Automatic Visualization

TLDR

AutoVis is a data viewer that automatically selects appropriate visualizations for diverse data types—content text, relational tables, hierarchies, streams, and images—using grammar of graphics, scagnostics, and statistical logic, thereby distinguishing it from generic automated chart‑generation tools. An AVS is intended to give researchers a quick, anomaly‑aware overview of their data before modeling, shielding them from missing values, outliers, miscodes, and other issues that could invalidate statistical assumptions. The system combines a drag‑and‑drop interface, a graphics generator that produces graphs without user definitions, a statistical analyzer that prevents false conclusions, and a pattern recognizer that identifies density, shape, trend, and other cues that professional statisticians look for.

Abstract

AutoVis is a data viewer that responds to content–text, relational tables, hierarchies, streams, images–and displays the information appropriately (that is, as an expert would). Its design rests on the grammar of graphics, scagnostics and a modeler based on the logic of statistical analysis. We distinguish an automatic visualization system (AVS) from an automated visualization system. The former automatically makes decisions about what is to be visualized. The latter is a programming system for automating the production of charts, graphs and visualizations. An AVS is designed to provide a first glance at data before modeling and analysis are done. AVS is designed to protect researchers from ignoring missing data, outliers, miscodes and other anomalies that can violate statistical assumptions or otherwise jeopardize the validity of models. The design of this system incorporates several unique features: (1) a spare interface–analysts simply drag a data source into an empty window, (2) a graphics generator that requires no user definitions to produce graphs, (3) a statistical analyzer that protects users from false conclusions, and (4) a pattern recognizer that responds to the aspects (density, shape, trend, and so on) that professional statisticians notice when investigating data sets.