Lets say I have 3 different methods. I am trying to find the best technique to represent similarities (or differences) among these methods.

One way I thought of is to create a table and create my own correlation methodology (e.g., high, medium, and low). However, I am not sure if this is the best way to do it.

enter image description here

Here is the corresponding LaTeX code I created (just in case):


             & Method 1 & Method 2 & Method 3 \\
    Method 1 & High & High  & Low \\
    Method 2 & Medium & Low & Low \\
    Method 3 & Low & Low  & Low \\
\caption{Level of correlation among different methods}
  • Is it mathematics? Do the high and lows represent some numbers?
    – Kay
    Commented Mar 16, 2018 at 11:45
  • No not necessarily. It is general methods, for example methods to clean the house method 1 (vacuum) method 2 (broom) etc. And the similarity (high medium low) represents similarity. Commented Mar 16, 2018 at 11:47
  • In that sense, many factor will come into account. Time, Energy, Man Power, Cost etc etc... So these things should also be reflected in your representation.... By the way, do you have taken all these into account and accordingly saying High, low or medium...?
    – Kay
    Commented Mar 16, 2018 at 11:52
  • For simplicity, I assumed that my correlation methodology of high, medium, and low could represent an average representation of these metrics. My issue is my way of representing it is the best or their could be other, I was reading about heatmaps but I don't know if it could be the way. Commented Mar 16, 2018 at 11:54

2 Answers 2


I assume that the high, low and medium corresponds to some analytical values and are dependent on various factors. So how about to assign some values to all constraints and so to this high, low and mediums also. Then finally show as an shaded matrix plot e.g., enter image description here

  • Thank! what tool did you use to create this ? Commented Mar 16, 2018 at 13:43
  • You can do it in Matlab, or GNU Plot or GIS ... Here is an example of GNU Plot and that to for heatmaps also gnuplot.sourceforge.net/demo/heatmaps.html
    – Kay
    Commented Mar 16, 2018 at 14:52
  • 2
    I'm sure color-blind readers love reading your papers. Or people who print your paper in B&W.
    – user9646
    Commented Mar 16, 2018 at 17:40

I suggest that you put characterizing features in the rows or columns of your table. For the descriptions, you can put simple explanation to the advantage or drawback of the $feature_i$ in $method_j$. Something like, faster timer, or heavy memory usage. It depends on the field.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .