A Privacy-Preserving Distributed Contextual Federated Online Learning Framework with Big Data Support in Social Recommender Systems

Zhou, Pan; Wang, Kehao; Guo, Linke; Gong, Shimin; Zheng, Bolong

doi:10.1109/tkde.2019.2936565

Cited by 46 publications

(23 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Remark 2: Corollary 1 implies that if the non-private version of any private algorithm is known, then the expected regret for the private algorithm is bounded by (18).…”

Section: Theoretical Resultsmentioning

confidence: 99%

Differential Privacy in Social Networks Using Multi-Armed Bandit

Odeyomi

2022

IEEE Access

View full text Add to dashboard Cite

There has been an exponential growth over the years in the number of users connected to social networks. This has spurred research interest in social networks to ensure the privacy of users. From a theoretical standpoint, a social network is modeled as a directed graph network and interactions among agents in the directed graph network can be analyzed with non-Bayesian learning and online learning strategies. The goal of the agents is to learn the underlying time-varying true state of the network through repeated cooperative interaction among themselves. To overcome privacy challenges in social networks, recent research works include differential privacy in the social network analysis to guarantee the privacy of shared information among the agents. However, the common online learning strategy adopted in most existing work is the stochastic multi-armed bandit approach which assumes that the loss distribution is independent and identically distributed. This does not account for the arbitrariness of the time-varying true state in the social network. Therefore, this paper proposes a tougher but realistic setting that removes the restriction on the loss distribution. Two non-stochastic multi-armed bandit algorithms are proposed. The first algorithm uses the Laplace mechanism to guarantee differential privacy against a third-party intruder. The second algorithm uses the Laplace mechanism to guarantee differential privacy against both a third-party intruder and any spying agent in the network. The simulation results show that the agents' beliefs converge to the most dominant true state among the sequence of arbitrarily time-varying true states over the time horizon. The speed of convergence comes as a trade-off with privacy. Regret bounds are obtained for the proposed algorithms and compared to the non-private algorithm in the literature.

show abstract

“…Remark 2: Corollary 1 implies that if the non-private version of any private algorithm is known, then the expected regret for the private algorithm is bounded by (18).…”

Section: Theoretical Resultsmentioning

confidence: 99%

Differential Privacy in Social Networks Using Multi-Armed Bandit

Odeyomi

2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The study shows that FL is able to improve the efficiency and accuracy of imitation learning at a robot by using the knowledge of other robots. Zhou et al [35] propose a FL framework for social recommender systems. FL is used to learn a centralized model using the collaboration between a large number of clients with contextawareness and big data supports.…”

Section: A Application Of Fl For Wireless Iotmentioning

confidence: 99%

Federated Learning for Vehicular Internet of Things: Recent Advances and Open Issues

Yoshinaga

et al. 2020

IEEE Open J. Comput. Soc.

239

106

View full text Add to dashboard Cite

show abstract

“…The present study on BD’S RS is divided into three sections: Propose an analytical method for a specific type of Zhou et al [ 31 ] proposed an especially distributed federal contextual framework for online learning. This framework is supported by BD technology and is a social RS with privacy.…”

Section: Related Workmentioning

confidence: 99%

“…Organizations attach different importance to different types of data challenges. For example, in the health industry, the most important challenge for BD management is privacy [ 31 ]. There are several challenges related to data management, which are categorized into seven areas such as privacy, security, data and information sharing, cost/operational costs, data governance, and data ownership [ 30 ].…”

Section: Related Workmentioning

confidence: 99%

An integrated model for evaluation of big data challenges and analytical methods in recommender systems

Asemi

Kő

et al. 2022

J Big Data

View full text Add to dashboard Cite

The study aimed to present an integrated model for evaluation of big data (BD) challenges and analytical methods in recommender systems (RSs). The proposed model used fuzzy multi-criteria decision making (MCDM) which is a human judgment-based method for weighting of RSs’ properties. Human judgment is associated with uncertainty and gray information. We used fuzzy techniques to integrate, summarize, and calculate quality value judgment distances. Then, two fuzzy inference systems (FIS) are implemented for scoring BD challenges and data analytical methods in different RSs. In experimental testing of the proposed model, A correlation coefficient (CC) analysis is conducted to test the relationship between a BD challenge evaluation for a collaborative filtering-based RS and the results of fuzzy inference systems. The result shows the ability of the proposed model to evaluate the BD properties in RSs. Future studies may improve FIS by providing rules for evaluating BD tools.

show abstract

A Privacy-Preserving Distributed Contextual Federated Online Learning Framework with Big Data Support in Social Recommender Systems

Cited by 46 publications

References 36 publications

Differential Privacy in Social Networks Using Multi-Armed Bandit

Differential Privacy in Social Networks Using Multi-Armed Bandit

Federated Learning for Vehicular Internet of Things: Recent Advances and Open Issues

An integrated model for evaluation of big data challenges and analytical methods in recommender systems

Contact Info

Product

Resources

About