While a plethora of Privacy-Preserving Cloud Big Data Analytics models and paradigms have been proposed in literature, computational overheads still remain an important limitation at today. With the goal of addressing this relevant gap, in this paper we propose an innovative framework for supporting privacy-preserving Cloud big data analytics via the application of decentralized SOLID Personal Online Datastores (PODs), by specifically considering the case of multiple-Cloud environments . Indeed, PODs ensure the privacy of user data across multiple, different (networked) systems. The final aim is that of reducing the execution of computationally-expensive anonymization algorithms at the node , while improving throughput and scalability of the target system. A comprehensive experimental assessment and analysis session confirms the benefits coming from our research.

This paper focuses the attention on privacy-preserving identity and access management in multiple Cloud environments, which is an annoying problem in the modern big data era. Within this conceptual context, the paper describes contemporaneous models and issues, and put the basis for future solid solutions. Finally, we provide a summary table where we embed an innovative taxonomy of state-of-the-art research proposals in the reference scientific field.

In this paper, we introduce and describe in details an innovative framework that combines high-level identity verification policies, Zero-Knowledge proofprotocols and SOLID Decentralized Secure Data Stores, to support privacy-preserving identity management over Multi-Cloud Environments. We provide both UML modelling of the proposed methodology, as well as the software component blueprint of the proposed framework.

In this paper, we propose a general methodology for estimating the degree of the attribution methods precision and generality in machine learning interpretability. Additionally, we propose a technique to measure the attribution consistency between two attribution methods. In our experiments, we focus on the two well-known model agnostic attribution methods, SHAP and LIME, then we evaluate them on two real applications in the attack detection field. Our proposed methodology highlights the fact that both LIME and SHAP are lacking precision, generality, and consistency. Therefore, more inspection is needed in the attribution research field.