An ensemble learning model based on differentially private decision tree

Abstract Using differential privacy to provide protection for classification algorithms has become a research hotspot in data mining. In this paper, we analyze the defects differentially private decision tree named Maxtree, and propose an improved model DPtree. DPtree can use Fayyad theorem process continuous features quickly, adjust budget adaptively according sample category distributions leaf nodes. Moreover, overcome inevitable decline of ability trees, ensemble learning DPtree, namely En-DPtree. voting En-DPtree, multi-population quantum genetic algorithm, introduce immigration operators elite groups search optimal weights base classifiers. Experiments show that performance is better than En-DPtree always superior other competitive algorithms.