Data Visualization
The actual visualization is as important as the data that is being visualized, obviously, since it is the end product of the process. Thus, paying close attention to creating the best possible visualization for the data at hand is crucial.
Interactive visualizations have multiple elements/parts. Let's take a closer look at each element to understand what can go wrong and how to prevent such mistakes.
Choosing a Visualization
Once your data has been cleaned and prepared, and the features that you want to visualize have been chosen, the first step in creating a visualization is selecting the graph or plot that is going to display your data. This decision impacts the efficiency and ease with which your visualization can explain your data, and thus you need to ensure that you're picking a visualization that can accurately explain and describe your data.
In the previous chapters, we looked at three types of data – stratified, temporal...