Umeå University's logo

Interactions among different elements of complex networks are organized in a structured manner. The collective behavior of the elements of these networks is organized according to community structure. Several methods have been defined to automatically detect these substructures in the field known as community discovery. Most of the methods have been applied to static or aggregated data. Recently the identification of evolving communities has gained more attention. Studying the relations among individuals yields insights on how communities form and evolve, but there are some limits that should be enforced to respect individuals’ privacy while sharing and collecting their data. Privacy-protection techniques have been commonly applied to static data, while there are few methods that work on dynamic data. Recently, there have been some approaches to protect dynamic graphs with local edge-differential privacy that have been tested for community discovery applications. However, the evolution of the communities over time has not been evaluated on the privacy-protected data. We test the utility considering community discovery and evolution in time-varying networks for such localedge-ϵ-differential privacy methods. We show empirically how these algorithms can provide privacy while preserving the community lifecycles, for their privacy-aware study.

Differential privacy allows to publish graph statistics in a way that protects individual privacy while stillallowing meaningful insights to be derived from the data. The centralized privacy model of differential privacyassumes that there is a trusted data curator, while the local model does not require such a trusted authority.Local differential privacy is commonly achieved through randomized response (RR) mechanisms. This doesnot preserve the sparseness of the graphs. As most of the real-world graphs are sparse and have several nodes,this is a drawback of RR-based mechanisms, in terms of computational efficiency and accuracy. We thus,propose a comparative analysis through experimental analysis and discussion, to compute statistics with localdifferential privacy, where, it is shown that preserving the sparseness of the original graphs is the key factorto gain that balance between utility and privacy. We perform several experiments to test the utility of theprotected graphs in terms of several sub-graph counting i.e. triangle, and star counting and other statistics. Weshow that the sparseness preserving algorithm gives comparable or better results in comparison to the otherstate of the art methods and improves computational efficiency.

In today's networked systems a massive amount of data is produced every day. These data can be modelled using graphs, where the nodes typically correspond to users or devices and the edges to the connections between them. Almost all networks change over time, with new nodes and edges appearing or disappearing as the system matures. Therefore dynamic graph models are more adequate to analyse such networks than static graphs, and appropriate tools need to be implemented to protect them. In this paper we obtain an edge differentially private version of dynamic stochastic block model. We show experimentally that the trends in the dynamic stochastic block model obtained from the original data are well preserved with the additional privacy guarantees.

Advancements in machine learning and data science deal with the collection of a tremendous amount of data for research and analysis, following which there is a growing awareness among a large number of users about their sensitive data, and hence privacy protection has seen significant growth. Differential privacy is one of the most popular techniques to ensure data protection. However, it has two major issues: first, utility-privacy tradeoff, where users are asked to choose between them; second, the real-time implementation of such a system on high-dimensional data is missing. In this work, we propose BUDS+, a novel differential privacy framework that achieves an impressive privacy budget of 0.03. It introduces iterative shuffling, embedding for data encoding, converger function into a novel comparison system to converge the privacy threshold among the aggregated differentially private and noisy reports to further minimize the attack model's time.

Huge amounts of data are generated and shared in social networks and other network topologies. This raises privacy concerns when such data is not protected from leaking sensitive or personal information. Network topologies are commonly modeled through static graphs. Nevertheless, dynamic graphs better capture the temporal evolution and properties of such networks. Several differentially private mechanisms have been proposed for static graph data mining, but at the moment there are no such algorithms for dynamic data protection and mining. So, we propose two locally ϵ-differentially private methods for dynamic graph protection based on edge addition and deletion through the application of the noise-graph mechanism. We apply these methods to real-life datasets and show promising results preserving graph statistics for applications in community detection in time-varying networks.

The main contributions of this work are: extending the definition of local differential privacy for edges to the dynamic graph domain, and showing that the community structure of the protected graphs is well preserved for suitable privacy parameters.

Federated Learning or FL is the orchestration of centrally connected devices where a pre-trained machine learning model is sent to the devices and the devices train the machine learning model with their own data, individually. Though the data is not being stored in a central database the framework is still prone to data leakage or privacy breach. There are several different privacy attacks on FL such as, membership inference attack, gradient inversion attack, data poisoning attack, backdoor attack, deep learning from gradients attack (DLG). So far different technologies such as differential privacy, secure multi party computation, homomorphic encryption, k-anonymity etc. have been used to tackle the privacy breach. Nevertheless, there is very little exploration on the privacy by design approach and the analysis of the underlying network structure of the seemingly unrelated FL network. Here we are proposing the ΔDSFL framework, where the server is being decoupled into server and an an alyst. Also, in the learning process, ΔDSFL will learn the spatio information from the community detection, and then from DLG attack. Using the knowledge from both the algorithms, ΔDSFL will improve itself. We experimented on three different datasets (geolife trajectory, cora, citeseer) with satisfactory results.