DATAVIZ: How to hierarchize data

When you want to show a hierarchal structure in data and show proportions, some great charts can help.

How to hierarchize data 2020/06/23 | DATAVISUALIZATION

We propose five charts that can easily fit your need. Most of them are perfectly adapted for numeric data but we suggest one chart specially dedicated for string and text data types which are precious information more and more analyzed.

Some examples?

In customer knowledge approach, this unstructured data really helps to understand and have insights from tweets or client’s reviews analysis.

In industry, you have also more and more digital activities where operational teams spend time to fill information and comments on process that are mine gold when you aim to improve quality level or efficiency for exemple.

Circle Packing: Beautiful, efficient but space greedy chart

The circle packing is also known as a Circular Treepmap because it’s a variation of the famous Treemap with circles instead of rectangle.

The main advantage of circle packing is beauty: this chart gives a very esthetical representation of large dataset and has a better visual efficiency than common treemap. The difficulty is that is quite greedy on space and if you need to be compact, forget it! Each level of the hierarchy is represented by a group of circles. You can use the circle area for another dimension of your data and give one more attribute thanks to the color. The result is often amazing, and we advise to keep the circle packing chart for special and impactful datavisualization shows!

Sunburst Diagram: Powerful revisited version of the doughnut chart

The Sunburst Diagram has various other names like Ring chart, Multi-level Pie Chart, Sunburst partition. It’s a quite simple graphics: it is built on the common Doughnut chart but with multiple level depicted by concentric cercles. The root node of the hierarchy is the cercle in the center and each ring represents a hierarchy level. The slices are divided based on their hierarchical relationship to the parent slice. Angle are proportional to a dimension of the data. The color highlights group or specific category. Sunburst Diagram can be easily realized from common tool like Drop to Kibana. We recommend it to make synthetic and smart datavisualization.


Nightingale Rose Chart: the most influential visualization in the history

Nightingale Rose Chart (also known as Coxcomb Chart or Polar Area Diagram) is quite like a usual pie chart, except sectors are equal angles and differ rather in how far each sector extends from the center of the circle. This chart is probably one of the most influential visualizations of all time because of the larger area size of the outer segments that gives more emphasis. Florence Nightingale, nurse and statistician famously used such a visualization to display data on how many soldiers died in hospitals during the Crimean War.

Nightingale Rose chart

Hyperbolic Tree: the Scientifics favorite chart

Hyperbolic Tree is largely inspired by hyperbolic geometry and now I guess you are a beat worry on what we can explain on this topic; but keep cool. in few words, hyperbolic tree provides an intuitive way to display hierarchically organized information like biological classifications. The challenge is the number of the node and the space: Your chart may quickly suffer from visual clutter as the number of nodes per level can grow exponentially. The hyperbolic Tree is more dedicated for academics that aim to provide a very detailed hierarchy and usually use dynamic charts to make a interactive and comfortable datavisualization.

Hyperbolic Tree
Hyperbolic Tree regarding Big linked cancer data published by Muhammad Saleem

Word Cloud: the popular chart that makes love dataviz

Word cloud is also called Tag Cloud and is a perfect method to display how frequently words appears in a dataset. The size of the words is proportional to the frequency and the color can represent another dimension or just be used for esthetical purpose. Wordcloud is useful for quickly perceiving the most prominent terms: it is widely used in media and well understood by the public. It’s a good user-friendly datavisualization to reveal the potential of data like tweets or comments in an intuitive and often nice for a non-technical public.

Word Cloud