Big Data Visualization: Techniques and Applications

Data visualization has become a core tool in the era of big data, which significantly improves the efficiency of information transfer and decision support by transforming complex data into intuitive graphics. This paper systematically discusses the visualization methods of temporal data, proportional data and relational data, combines traditional technology and modern innovation, and deeply analyzes the application scenarios and challenges in different fields. First, for temporal data, this paper compares the visualization strategies for continuous and discrete data, proposes advanced methods (e.g., spiral diagrams and heat maps) suitable for dynamic analysis, and emphasizes the importance of real-time data processing. Second, in proportional data visualization, this paper analyzes the limitations of pie charts and stacked bar charts, introduces tools such as tree diagrams and sank diagrams to cope with high dimensionality and real-time demands, and analytically demonstrates the potential of AI-driven automated recommender systems. In the relational data section, the paper extends the application scenarios of scatterplot and heatmap, and combines radar and sank charts to solve the difficulties of multivariate interaction and streaming data mapping. In addition, the paper verifies the practical value of these techniques through practical applications in medical, financial, and e-commerce fields. Finally, the paper looks forward to future research directions, including augmented reality (AR) visualization, edge computing-supported low-latency analytics and optimization of ethical design principles. This research provides practical guidelines and innovations for big data visualization.