Options

Interpreting Correlation

k_vishnu772k_vishnu772 Member Posts: 34 Contributor I
edited November 2018 in Help

HI All ,

 

I  need small help in interpreting the results of cross correlation.i am using it for classification problem where my target is a binary (yes or no) and most of my input variables are polynomials and i ran the correlation after converting them to numerical variables .lets take one example where i have variable1(V1,V2,V4,V5) and variable2 (T1,T2,T3,T5) and i got the correlation result of V1 and T1 as 0.345 and i don't understand how to interpret this value in relation to my binary target.Could ou please help .

 

Regards,

Vishnu

Answers

  • Options
    Telcontar120Telcontar120 Moderator, RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,635 Unicorn

    You shouldn't interpret the correlation coefficent at all for nominal variables!  I believe the internal calculation is just the correlation based on ranked category values, but this is rarely meaningful.

    Brian T.
    Lindon Ventures 
    Data Science Consulting from Certified RapidMiner Experts
  • Options
    SGolbertSGolbert RapidMiner Certified Analyst, Member Posts: 344 Unicorn

    I agree with Brian. However, you could use the techniques of Polychorical Correlation or Correspondence Analysis to analize your tables. They are both descriptive statistical processes, which are not included in RM.

     

    Regards,

    Sebastian

Sign In or Register to comment.