Title: Efficient and Private Scoring of Decision Trees, based on Pre-Computation Technique with Support Vector Machines and Logistic Regression Model
Abstract: Numerousinformation driven customized administrations require that private information of clients is scored against a prepared machine learning model.In this paper we propose a novel convention for security protecting order of choice trees, a famous machine learning model in these situations.Our answers is made out of building squares, to be specific a safe correlation convention, a convention for negligently choosing inputs, and a convention for increase.By joining a portion of the building hinders for our choice tree order convention, we additionally enhance beforehand proposed answers for characterization of help vector machines and calculated relapse models.Our conventions are data hypothetically secure and, dissimilar to already proposed arrangements, don't require secluded exponentiations.We demonstrate that our conventions for protection saving arrangement prompt more proficient outcomes from the perspective of computational and correspondence complexities.We introduce exactness and runtime comes about for 7 characterization benchmark datasets from the UCI archive.