The RapidMiner community is on read-only mode until further notice. Technical support via cases will continue to work as is. For any urgent licensing related requests from Students/Faculty members, please use the Altair academic forum here.
"Introducing Missing Values"
Hi!
How can I mark some value as missing?
I have some data where each feature (real-valued) of each instance has a value. Now I have mark some values as missing. How can I do this in RM?
Best regards,
chero
How can I mark some value as missing?
I have some data where each feature (real-valued) of each instance has a value. Now I have mark some values as missing. How can I do this in RM?
Best regards,
chero
Tagged:
0
Answers
this depends when you want to change the data? Before importing the data into RapidMiner or inside a RapidMiner process?
Greetings,
Sebastian
I want to change it inside a RM process.
Best regards,
chero
I would use the Generate Attributes operator for this. You might define conditions there to decide when a value should become unknown. Unfortunately this isn't encodable directly, but you could enter 0/0 to define a value as missing. Here's a sample process: Greetings,
Sebastian
this is no solution to my problem. I use some program to encode specific data. This program doesn't allow to set some data as missing furthermore there is no value which i can use to mark some value as missing. The resulting data is written in a specific format.
I wrote my own RM operator to import this data. Now I have to say something like this: attribute 1 of example 13 is missing. This is NOT codeable in the data.
The only solution I see is importing it into RM and exporting it to csv (or similar). In that file I could give those attributes specific values which I could recode in RM the way you suggested. Is there an easier way.
Best regards,
chero
Since you have written your own input operator, setting missing values should be piece of cake for you
steffen
I haven't tried it yet. But it seams to be the easiest solution. I'll give it a try in the next days.
Best regards,
chero