Weights provided by Correlation Matrix

qwertz

Dear all,

first let me reference to another post where I initially discovered this problem. http://rapid-i.com/rapidforum/index.php/topic,5383.0.html
Anyway, I thought that this is something different from what was the original question in the other post. Furthermore, I think that this very problem is better suited in the category "problems and suppor". That's why I open another thread.



Regarding my observations:

The grid of the correlation matrix shows the correlation coefficient between two attributes. However, the weights provided by this operator appear to be illogical to me.


          a1    a2    a3
a1  1,00  0,35  0,73
a2  0,35  1,00  0,11
a3  0,73  0,11  1,00


To calculate the weights it sums up all values of a row (e.g. 1 + 0,35 + 0,73 = 2,08)
Then this sum is being devided by the number of attributes (2,08 / 3 = 0,693)
This result is then substracted from 1 (1 - 0,693 = 0,307)


At first I am wondering that the correlation of an attribute to itself is being considered (which is always 1).

Furthermore, by substracting from 1 the weights become higher for those attributes that show a lower correlation. This is not done by the "weight by correlation" operator.


Bye for now
Sachs