Privacy Preserving Ensemble Learning Classification Model for Mental Healthcare

Abstract

Privacy protection technology can effectively handle personal data to prevent the leakage of sensitive information during data publishing and data analysis. Countless research analysis on facilitating privacy-preserving data collection has been conducted through the years. Medical research represents an important application that is necessary to extract useful information and protect patient privacy One of the main techniques recently used is ensemble learning. In contrast to ordinary approaches, which try to learn one hypothesis from a training dataset, ensemble learning tries to construct a set of hypotheses and combine them. This strategy is primarily used to improve the performance of the model, but the major restraint of this approach is that multiple parties cannot share their data extracted from the ensemble learning model with a privacy guarantee; therefore, it is a great demand to develop privacy-preserving collaborative ensemble learning. In this paper, introducing a differential privacy-based ensemble learning classification model while maintaining its utility and accuracy, which will evaluate the medical health record and classify the patient’s medical illness from the given data in a way that requires very little human participation from doctors. The proposed model has lower time and money costs associated and will provide assistance to medical practitioners.

Ähnliche Arbeiten