What are some ways of making a graph data look better?

Question

---

I'm not quite sure if this is the correct place to post this, as it's not strictly mathematics.

---

I have an example dataset:

 User   Value  
------ ------- 
  A        100  
  B        200  
  C       1000  
  D      19000  

I'm running into a problem graphing the data, when I have such a huge distance between the numbers, the graph 'looks' odd - the values are too far apart which leads to aesthetically displeasing graphs.

Essentially, I'm trying to make the graphs 'less accurate', but better looking. Can I put the values into through an algorithm that maybe shifts everything that looks like it's outside of some calculated pivot point?

Sorry if I'm not super clear.

Thanks,

Ollie

Answer 1

If the range of one of the variables spans several orders of magnitude a good you can take log (or equivalently, plot in a log-scale), here's an example in python

import matplotlib.pyplot as plt
import numpy as np

labels = ['A', 'B', 'C', 'D']
x = [1, 2, 3, 4]
y = [100, 200, 1000, 19000]

plt.plot(x, np.log(y), 'or')
plt.xticks(x, labels)
plt.show()