RapidMiner

How to calculate multiple correlation matrices by looping through subset (based on feature values)

Contributor

How to calculate multiple correlation matrices by looping through subset (based on feature values)

Dear community,

 

I have a data set that contains student names, their marks in different courses and the number of times they showed up for classes, like e.g.:

 

Max Mustermann | Database Technology | 2.3 | 10

Max Mustermann | Computer Science Intro | 1.3 | 12

Maria Musterfrau | Computer Science Intro | 2.0 | 13

...

 

So Max showed up 10 times at DB and got a 2.3 (mark) in the end.

 

Now, for each student in the data set, I want to calculate the correlation coefficient between mark and number of times attended.

 

The output should look like this:

 

Student name | Correlationcoefficient between mark and number of times attended (considering all courses the respective student took)

Max Mustermann | 0,711

Maria Musterfrau | -0,312

...

 

Everything works fine, except that I don't know how to best do the looping and producing the final table (e.g. using "correlation matrix").

 

I would appreciate your help. Thanks in advance and kind regards.

See more topics labeled with:

1 REPLY
Community Manager

Re: How to calculate multiple correlation matrices by looping through subset (based on feature value

So your loop calculates the average and # of attendance for each student?  Then take the resulting exampleset and connect a correlation matrix operator to it.Out put the MAT port

Regards,
Thomas
LinkedIn: Thomas Ott
Blog: Neural Market Trends