How Data Visualization Shapes Big Data Analysis: Evolution, Techniques, and Trends

In the digital era, data visualization has become essential for interpreting the massive amounts of information generated by big‑data applications. The article first outlines the historical development of visualization, noting that before the 17th century there were no line or bar charts, that the 18th century introduced basic bar and time‑series graphs, and that the widespread adoption of computers in the late 1960s enabled the transition from hand‑drawn graphics to programmatic visualizations. Today, the focus has shifted to dynamic, interactive visualizations that can handle high‑dimensional data.

Key Challenges

Perceptual interaction: large‑scale and high‑dimensional datasets cause latency and reduce interactivity.

User‑driven drawing: encouraging users to imagine data before it is visualized.

Multi‑device presentation: adapting visualizations designed for large screens to mobile devices.

Major Visualization Categories

Text visualization : Uses natural‑language processing (e.g., tokenization, named‑entity recognition, keyword extraction, topic modeling, sentiment analysis) to transform textual data into visual forms such as word clouds, topic maps, and relational diagrams.

Network visualization : Represents entities as nodes and relationships as edges, revealing structure, communities, and diffusion processes.

Spatio‑temporal visualization : Handles data that varies over space and time, supporting static and dynamic displays for queries, pattern mining, and trajectory clustering.

Multidimensional visualization : Reduces high‑dimensional data to comprehensible graphics using facets, color, shape, size, and animation to expose inter‑dimensional relationships.

Visualization Schools of Thought

Two main schools are discussed:

Chart taxonomy (type‑based): Classifies visualizations by familiar chart names such as bar, line, pie, radar, funnel, waterfall, etc. Tools like Power BI and Tableau illustrate this approach by letting users select a chart type and then configure axes, legends, and styling.

Graphic grammar (grammar‑based): Defines a formal language for describing visualizations. The grammar consists of six concepts: DATA (variables from datasets), TRANS (transformations like sorting), SCALE (e.g., logarithmic scaling), COORD (coordinate systems such as Cartesian or polar), ELEMENT (visual marks with channels like color and size), and GUIDE (auxiliary elements like axes and legends). The process follows three stages – “specify”, “assemble”, and “display”.

Core Visualization Technologies

OpenGL : A cross‑platform 2D/3D graphics API widely used for games, industrial modeling, and embedded devices. Since version 2.0 (2004) it supports GLSL shaders, and later versions have removed deprecated fixed‑function pipelines in favor of programmable pipelines.

WebGL : Brings GPU‑accelerated 3D rendering to browsers via JavaScript, enabling complex visualizations without plugins. It is supported by major desktop and mobile browsers and underpins many modern data‑visualization libraries.

Canvas : An HTML5 element that provides an immediate‑mode drawing API for 2D graphics. It offers high performance for pixel‑level operations but can suffer from performance bottlenecks with very large scenes.

SVG : An XML‑based vector graphics format that integrates with HTML/CSS. It excels at resolution‑independent graphics and declarative styling, though complex SVG trees can impact performance.

Future Outlook

With the rise of 5G, IoT, cloud computing, and AI, visualization is moving toward real‑time, multi‑device, and AI‑augmented experiences. Researchers are exploring deeper integration of GPU‑based rendering, hybrid server‑client pipelines, and novel visual grammars to handle ever‑growing data volumes and complexity.

References

Jia Q., Chai C., Cai R. "Design Aesthetics in Data Visualization" Packaging Engineering, 2022.

Zheng Y., Zhao Y., Bai X., et al. "Survey of Educational Big Data Visualization" Computer Science & Exploration, 2021.

Chen W., Shen Z., Tao Y., et al. Data Visualization , Beijing: Publishing House of Electronics Industry, 2019.

Lei W. "Research on the Development of Data Visualization" Electronic Technology & Software Engineering, 2017.

Shawn Shen. "Design of BI Chart Visualization Editor" Zhihu, 2019.

Lan Xingyu, Wang Jiazhe. "Practices of Chart Taxonomy in Data Visualization Design", Art Review, 2022.

Yang L. "Research on Data Visualization Based on Graphic Grammar", Computer Knowledge & Technology, 2020.

Gordon V. S., Clevenger J. "Computer Graphics Programming with OpenGL and C++".

Mo Z. "From 0 to 1: HTML5 Canvas Animation Development" People's Posts and Telecommunications Press, 2020.