What Is Data Visualization?

Data visualization is the practice of representing data graphically to communicate information, reveal patterns, and support analysis. It translates quantitative or relational data into visual forms, including charts, maps, trees, and network diagrams, so that human perception can detect structure that would be difficult to extract from raw numbers or text. Effective visualization is not merely decorative: it is a primary means of hypothesis generation in scientific research, a tool for executive decision-making in business intelligence, and an essential step in exploratory data analysis before any formal modeling begins.

The field draws on statistics, computer graphics, cognitive psychology, and human-computer interaction. It distinguishes between scientific visualization, which renders spatial and physical simulation data, and information visualization, which represents abstract, non-spatial data such as financial records, software structures, or social networks. Both branches converge in the study of visual analytics, where interactive visualization is combined with computational analysis to support reasoning over large and complex datasets.

Scientific and Information Visualization

Scientific visualization focuses on data that has an inherent spatial structure, such as computational fluid dynamics simulations, medical imaging volumes, or atmospheric models. The goal is to render these fields in ways that expose their three-dimensional topology, gradients, and dynamics to the viewer. Techniques include volume rendering, isosurface extraction, and streamline tracing for vector fields. Information visualization, by contrast, operates on abstract data with no natural spatial embedding. Canonical methods include bar charts and scatter plots for low-dimensional numerical data, treemaps and sunburst diagrams for hierarchical data, node-link graphs for network relationships, and heat maps for matrix-valued comparisons. The IEEE Transactions on Visualization and Computer Graphics is the primary research journal for both branches, publishing work on new representation techniques, perceptual evaluation studies, and rendering algorithms.

Visual Analytics and Data Mining

Visual analytics combines interactive visualization with automated data mining and statistical modeling so that analysts can steer computational processes in real time. A visual analytics workflow typically begins with data mining algorithms (such as clustering, dimensionality reduction, or anomaly detection) that reduce a high-dimensional dataset to a set of candidate structures; the analyst then examines those structures visually, adjusts parameters, and re-runs the computation in a tightly coupled feedback loop. Dimensionality reduction methods including t-SNE and UMAP project high-dimensional data onto two or three dimensions in a way that preserves local neighborhood structure, making cluster separation visible. Tools such as D3.js for browser-based visualization and libraries in R and Python (ggplot2, Matplotlib, Seaborn, Plotly) implement these workflows across a spectrum of sophistication, from publication-quality static plots to fully interactive dashboards. Overview of Data Visualization methods published in PMC surveys techniques used across scientific domains.

Modeling and Encodings

The expressiveness of a visualization depends on the visual encodings chosen: how data attributes are mapped to position, size, color, shape, and orientation. Position on a common scale is the most accurately perceived encoding, which is why aligned bar charts support more precise comparisons than bubble area or color saturation. Color choice requires attention to perceptual uniformity (so that equal numerical differences appear as equal visual differences) and to colorblindness accessibility; the Viridis and ColorBrewer palettes were designed specifically with these constraints in mind. Animation and interactivity extend static encodings to temporal and exploratory contexts, allowing viewers to filter, zoom, and drill down into subsets of interest. The ColorBrewer guidance for cartographic data classification by Cynthia Brewer at Penn State provides a widely used reference for selecting color scales in quantitative maps and visualizations.

Applications

Data visualization has applications in a wide range of disciplines, including:

• Biomedical imaging and clinical diagnostics relying on volume rendering of MRI and CT data
• Financial market analysis using candlestick charts and time-series dashboards
• Genomics and bioinformatics employing heat maps and phylogenetic trees
• Network security monitoring through real-time traffic graphs and anomaly displays
• Urban planning and geospatial analysis using choropleth and dot-density maps
• Journalism and public communication through interactive news graphics