Beginners trying to understand RapidMiner and what functions to use

shosa77

Hi, hello and welcome!

 

So we have this school lab we need help with.

As part of the teacher training program development efforts, the program director is trying to understand if there are groups of students who share specific characteristics that are not obvious, in order to try to tailor the teaching approaches as best fits their needs. So the director asks you, the analyst, to help him with this task. During another project last year, his team has collected psychological data on 600 students, attributes such as locus of control, (academic) self-concept and motivation to learn. These psychological characteristics were combined with the students’ academic performance on four standardized tests: reading, writing, math and science. In addition, the dataset[1] includes the student’s gender. 

 

We have tried to serveral connection/functions like:

Read Excel -> Validation -> Testing -> Training. In training we use K-nn (lazy) and in testing we use Apply model and Performance(Classification).

 

So any ideas of what connections to use? What are we missing? Please guide us!

 

I have attached the file we are using.

Sorry if we are in the wrong segment or subforum.

MartinLiebig

Dear Shosa,

 

thanks for reaching out. It sounds like your way is a good start. Have you had a look on Normalize as well as other models? A decision tree is usually a interesting next step

 

~Martin

shosa77

Sorry, the file went missing. But here it is.

MartinLiebig

Dear Shosa,

 

thanks for reaching out. It sounds like your way is a good start. Have you had a look on Normalize as well as other models? A decision tree is usually a interesting next step

 

~Martin

shosa77

Hi Martin!

 

Thank you for the quick reply.

 

We will look into the decision tree and see if we can get our heads around it!