Helping predictive analytics interpretation using regression trees and clustering perturbation

摘要: Regression trees are helpful tools for decision support and predictive analytics, due to their simple structure the ease with which they can be obtained from data. Nonetheless, when applied non-trivial datasets, tend grow according complexity of data, becoming difficult interpret. This difficulty overcome by clustering dataset representing regression tree each cluster independently. In order help create models that more comprehensible, we propose in this work a perturbation method reduce size cluster. A prototype has been developed tested on several datasets.