As a result of the possible responses to a questionnaire, I have 30 different variables and 60 observations.

From these variables' values, I have 5 subsets of data that each of them consists of some variables and some observations. Each subset has its own label which is a variable that defines the type of the observations and in one subset there is only one type as the label.

Now I want to make a classification model for new observations, should I combine all these subsets together into 1 dataset for building the model?

if yes, the problem is that the variables in all subsets are different and not the same and as a consequence of merging the subsets lots of missing values will be created, what is the solution then?

If not should I build 5 different classification models? and when I have only one label for all observations of a subset of data, how should I develop such a model?

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.