Frequent Pattern Mining (FPM) - Privacy Preserving and Security of Intermediate Data in Cloud Storage

Abstract

Nowadays, cloud computing has played a vital role in most data-intensive applications to store the data in the intermediated dataset. This effective cloud storage process helps to minimize the storage and processing cost while performing recomputing. Although the cloud provides numerous services, resources maintaining cost, outsourced user data protection from unauthorized users, the privacy of sensitive data, and computation complexity is still a major issue. A novel network-based frequent pattern mining model (DLFPM) is introduced to overcome these issues. Here, the presented method examines the frequent access information according to the layers of network functions. The network computes the sensitive information from the deep learning function. The identified sensitivity information is encrypted using the single sign-on associated with the Paillier encryption technique (SSO-PE) that avoids unauthorized access. The effective utilization of these algorithms continuously manages the sensitive data security that helps to minimize the computation cost and computation time.