A Survey on Privacy-Preserving Outsourced Data on Cloud with Multiple Data Providers

Abstract

Deep Learning has become the most common technique used in different fields like pattern recognition, weather prediction, medical diagnosis, banking sectors, aerospace and defense, data mining applications, image processing, speech recognition, cancer analysis, etc. The training of the model demands an enormous amount of data that is gathered from various data providers and sent to the cloud because they have limited storage and less computational resources. Cloud trains the model with the help of the gathered data. Also, data providers have genuine and confidential data that emerges the privacy issues, so to preserve privacy, the data is sent in the encrypted form and cannot be exposed to anyone and must be secured also. This paper focuses on the pros and cons of the different algorithms and techniques used for machine and deep learning models used for privacy preservation.