Web-based Information Visualization

It's unprofessional and unethical to do data analysis using only statistics. This talk shows why visualization is essential for revealing what the numbers hide.

#1about 8 minutes

Why information visualization is essential for data analysis

Visualizing data is crucial because humans interpret images better than numbers, as demonstrated by Anscombe's Quartet where identical statistics hide vastly different datasets.

#2about 6 minutes

The process of mapping data to visual attributes

Data mapping is the core process of converting data attributes into visual representations like position, size, shape, and color to create effective charts.

#3about 4 minutes

A survey of visualization tools and libraries

An overview of the visualization landscape covers fully-featured applications like Tableau and Power BI, as well as programming libraries for Python, R, and JavaScript.

#4about 6 minutes

Using SVG and Canvas for web visualizations

Web-based visualizations are primarily built using SVG for its vector graphics and native interactivity, while Canvas offers more complex drawing capabilities with WebGL.

#5about 4 minutes

A closer look at the D3.js library

D3.js is a powerful and flexible open-source JavaScript library that binds data to DOM elements, enabling a wide range of custom and interactive visualizations.

#6about 10 minutes

Choosing tools for analysis versus presentation

Data visualization serves different use cases like exploration and presentation, and a comparative study shows how tools like D3.js excel at flexible presentation.

#7about 5 minutes

Overcoming challenges of data size and security

Web-based visualizations must address performance issues with large datasets through aggregation and sampling, and handle the security implications of client-side data transfer.

#8about 1 minute

Key takeaways for effective web-based visualization

To succeed with web-based visualization, it is crucial to know your users and use case, carefully select the data to be shown, and leverage open-source libraries.