Algorithm for smoothing data values ​​for visualization

Question

I read some data for countries around the world and play with Google visualization gadgets, in particular map visualization . The problem is that the US is always ahead. While most countries have values ​​from 1 to 50, the United States consistently has a value of 2000+. This means that when rendering, it’s hard to tell the difference between all the “small countries,” since they all have the same shade of pale green, while the United States is always greasy dark green.

I don’t really care about the accuracy of the visualization, so I would like to smooth or average the values ​​a little so that there is a visible difference between a very low, low and not very low country. What is a good algorithm for this?

Pretty simple problem, but I'm not a mathematician at all. ^ _ ^ ;;

Answer 1

How to display data on a logarithmic scale. Thus, the value 10 is converted to 1, the value 100 is converted to 2, 1000 is converted to 3, etc.

Answer 2

Log scales are an option if, generally speaking, there is a magnitude of the order of the difference between most important data points.

However, if your distribution is bimodal, you better normalize and then record the fragmentation of your data.

To do this, you have to find a parameter that correlates with the 2000+ United States (possibly GDP)? and normalize all data points to the regional value of this number. You will see the exact differences that interest you in the intuitive color map.