Questions about attribute types
Contributor I
This is my first time using RapidMiner and I'm just learning data mining in general so some of the vocab is not clear.
I'm using a large database of direct mail info as my training set. My ultimate goal is to give each record a ranking and find a threshold that tells me they will or will not donate.
I started on this data set earlier today but I am starting over because I got a lot of errors when i tried running processes. More on that later.
Do I do the preparation of the data, stop, then start on the modeling? Because it seems to want to do everything together and that's a little scary.
I am still in the process of (re)importing the data, step 4 of the wizard. I'm actually on another computer so this is the first time it's seeing the data.
I'm un-checking most of the columns because I won't need them. I want to make sure i have the remaining columns set up correctly, because I didn't see an easy way to change individual columns from one type to another after the data was loaded. Please tell me if there's an EASY way.
I have a column that says whether they gave to the last mailing. I assumed this should be "label". Earlier an error said something like I must have a prediction column. Should this be my prediction column instead?
Any help is appreciated! I tried watching the videos but the music is too loud for me to hear what he's saying, with that accent.
I'm using a large database of direct mail info as my training set. My ultimate goal is to give each record a ranking and find a threshold that tells me they will or will not donate.
I started on this data set earlier today but I am starting over because I got a lot of errors when i tried running processes. More on that later.
Do I do the preparation of the data, stop, then start on the modeling? Because it seems to want to do everything together and that's a little scary.
I am still in the process of (re)importing the data, step 4 of the wizard. I'm actually on another computer so this is the first time it's seeing the data.
I'm un-checking most of the columns because I won't need them. I want to make sure i have the remaining columns set up correctly, because I didn't see an easy way to change individual columns from one type to another after the data was loaded. Please tell me if there's an EASY way.
I have a column that says whether they gave to the last mailing. I assumed this should be "label". Earlier an error said something like I must have a prediction column. Should this be my prediction column instead?
Any help is appreciated! I tried watching the videos but the music is too loud for me to hear what he's saying, with that accent.
0
Answers
these are a lot of questions.. I would recommend you to have a look at our video tutorial here: http://rapid-i.com/content/view/189/212/lang,en/
Furthermore you can have a look at the RapidMiner user manual which can be found here: http://rapid-i.com/content/view/36/210/lang,en/
Last you can have a look at the example processes found under Samples/processes in your RapidMiner Repositories View.
I think this should answer most of your questions.
Best,
Nils