<i>D</i>
            2
            <i>P</i>

Guerraoui, Rachid; Kermarrec, Anne-Marie; Patra, Rhicheek; Taziki, Mahsa

doi:10.14778/2757807.2757811

Cited by 31 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Typically, we are interested in values of δ that are smaller than the inverse of the database size. Although DP has been adopted to many domains [7] such as recommendation systems [31], we are not aware of any work besides [18] which adopts DP for SVD computation. Thus, as we later show a flaw in that work, we are the first to provide a distributed SVD computation with DP guarantees.…”

Section: Definition 1 ((ε δ)-Differential Privacy)mentioning

confidence: 99%

Privacy-Preserving Federated Singular Value Decomposition

2023

View full text Add to dashboard Cite

Singular value decomposition (SVD) is a fundamental technique widely used in various applications, such as recommendation systems and principal component analyses. In recent years, the need for privacy-preserving computations has been increasing constantly, which concerns SVD as well. Federated SVD has emerged as a promising approach that enables collaborative SVD computation without sharing raw data. However, existing federated approaches still need improvements regarding privacy guarantees and utility preservation. This paper moves a step further towards these directions: we propose two enhanced federated SVD schemes focusing on utility and privacy, respectively. Using a recommendation system use-case with real-world data, we demonstrate that our schemes outperform the state-of-the-art federated SVD solution. Our utility-enhanced scheme (utilizing secure aggregation) improves the final utility and the convergence speed by more than 2.5 times compared with the existing state-of-the-art approach. In contrast, our privacy-enhancing scheme (utilizing differential privacy) provides more robust privacy protection while improving the same aspect by more than 25%.

show abstract

Section: Definition 1 ((ε δ)-Differential Privacy)mentioning

confidence: 99%

Privacy-Preserving Federated Singular Value Decomposition

2023

View full text Add to dashboard Cite

show abstract

“…This new scheme was divided into four steps: (1) Constructing multiple attribute characteristic vector safety index for documents; (2) Constructing reverse index for the uploaded documents and generating vector set of document, then computing module of each document vector; (3) Encrypting document vector set with homomorphic encryption and uploading them into cloud; (4) Adopting multiple attribute score formula to calculate document relevance score, according to the scores ranking to return the interesting retrieval results for the users. Guerraoui et al (2015) presented a novel protocol distance-based differential privacy (D2P) that ensured a strong form of differential privacy, called distance-based differential privacy, and it was particularly well suited to recommenders. It combined random interference and differential privacy, a hybrid privacy protection recommendation system was proposed.…”

Section: Related Workmentioning

confidence: 99%

P3OI-MELSH: Privacy Protection Target Point of Interest Recommendation Algorithm Based on Multi-Exploring Locality Sensitive Hashing

et al. 2021

View full text Add to dashboard Cite

With the rapid development of social network, intelligent terminal and automatic positioning technology, location-based social network (LBSN) service has become an important and valuable application. Point of interest (POI) recommendation is an important content in LBSN, which aims to recommend new locations of interest for users. It can not only alleviate the information overload problem faced by users in the era of big data, improve user experience, but also help merchants quickly find target users and achieve accurate marketing. Most of the works are based on users' check-in history and social network data to model users' personalized preferences for interest points, and recommend interest points through collaborative filtering and other recommendation technologies. However, in the check-in history, the multi-source heterogeneous information (including the position, category, popularity, social, reviews) describes user activity from different aspects which hides people's life style and personal preference. However, the above methods do not fully consider these factors' combined action. Considering the data privacy, it is difficult for individuals to share data with others with similar preferences. In this paper, we propose a privacy protection point of interest recommendation algorithm based on multi-exploring locality sensitive hashing (LSH). This algorithm studies the POI recommendation problem under distributed system. This paper introduces a multi-exploring method to improve the LSH algorithm. On the one hand, it reduces the number of hash tables to decrease the memory overhead; On the other hand, the retrieval range on each hash table is increased to reduce the time retrieval overhead. Meanwhile, the retrieval quality is similar to the original algorithm. The proposed method uses modified LSH and homomorphic encryption technology to assist POI recommendation which can ensure the accuracy, privacy and efficiency of the recommendation algorithm, and it verifies feasibility through experiments on real data sets. In terms of root mean square error (RMSE), mean absolute error (MAE) and running time, the proposed method has a competitive advantage.

show abstract

Differential Privacy in Online Dating Recommendation Systems

Steiner

2020

Privacy and Identity Management. Data for Better Living: AI and Privacy

View full text Add to dashboard Cite

By their very nature, recommendation systems that are based on the analysis of personal data are prone to leak information about personal preferences. In online dating, that data might be highly personal. The goal of this work is to analyse, for different online dating recommendation systems from the literature, if differential privacy can be used to hide individual connections (for example, an expression of interest) in the data set from any other user on the platform -or an adversary that has access to the information of one or multiple users. We investigate two recommendation systems from the literature on their potential to be modified to satisfy differential privacy, in the sense that individual connections are hidden from anyone else on the platform. For Social Collab by Cai et al. we show that this is impossible, while for RECON by Pizzato et al. we give an algorithm that theoretically promises a good trade-off between accuracy and privacy. Further, we consider the problem of stochastic matching, which is used as the basis for some other recommendation systems. Here we show the possibility of a good accuracy and privacy trade-off under edge-differential privacy.

show abstract

D 2 P

Cited by 31 publications

References 22 publications

Privacy-Preserving Federated Singular Value Decomposition

Privacy-Preserving Federated Singular Value Decomposition

P3OI-MELSH: Privacy Protection Target Point of Interest Recommendation Algorithm Based on Multi-Exploring Locality Sensitive Hashing

Differential Privacy in Online Dating Recommendation Systems

Contact Info

Product

Resources

About