How to create an ensemble of 'Tree' learners from multiple data sources ?

I'm working on a project to classify network traffic of Applications, since each Application is used by many users therefore I'll have multiple data sets describing the same application from multiple users and this is the case of each application. I was able to classify the Apps for one user training a tree classifier from one data set. However my aim now is to create a classifier for each user apps and then ensemble the classifiers using bagging or boosting.

therefore is there a way to create this ensemble of classifiers using same classification (learning) method from multiple data sources ?

Thank you.