Privacy-preserving outsourced classification in cloud computing

Classifier has been widely applied in machine learning, such as pattern recognition, medical diagnosis, credit scoring, banking and weather prediction. Because of the limited local storage at user side, data and classifier has to be outsourced to cloud for storing and computing. However, due to privacy concerns, it is important to preserve the confidentiality of data and classifier in cloud computing because the cloud servers are usually untrusted. In this work, we propose a framework for privacy-preserving outsourced classification in cloud computing (POCC). Using POCC, an evaluator can securely train a classification model over the data encrypted with different public keys, which are outsourced from the multiple data providers. We prove that our scheme is secure in the semi-honest model