SMAP: A Joint Dimensionality Reduction Scheme for Secure Multi-Party Visualization

Jiazhi Xia; Tianxiang Chen; Lei Zhang; Wei Chen; Yang Chen; Xiaolong Zhang; Cong Xie; Tobias Schreck

doi:10.1109/VAST50239.2020.00015

SMAP: A Joint Dimensionality Reduction Scheme for Secure Multi-Party Visualization

Jiazhi Xia, Tianxiang Chen, Lei Zhang, Wei Chen, Yang Chen, Xiaolong Zhang, Cong Xie, Tobias Schreck

Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review

Abstract

Nowadays, as data becomes increasingly complex and distributed,
data analyses often involve several related datasets that are stored
on different servers and probably owned by different stakeholders.
While there is an emerging need to provide these stakeholders with
a full picture of their data under a global context, conventional
visual analytical methods, such as dimensionality reduction, could
expose data privacy when multi-party datasets are fused into a
single site to build point-level relationships. In this paper, we
reformulate the conventional t-SNE method from the single-site
mode into a secure distributed infrastructure. We present a secure
multi-party scheme for joint t-SNE computation, which can minimize the risk of data leakage. Aggregated visualization can be
optionally employed to hide disclosure of point-level relationships.
We build a prototype system based on our method, SMAP, to
support the organization, computation, and exploration of secure
joint embedding. We demonstrate the effectiveness of our approach with three case studies, one of which is based on the deployment of our system in real-world applications.

Original language	English
Title of host publication	Proceedings - 2020 IEEE Conference on Visual Analytics Science and Technology, VAST 2020
Pages	107-118
Number of pages	12
ISBN (Electronic)	978-1-7281-8009-0
DOIs	https://doi.org/10.1109/VAST50239.2020.00015
Publication status	Published - Oct 2020
Event	IEEE VIS 2020 - Virtuell, United States Duration: 25 Oct 2020 → 30 Oct 2020 http://ieeevis.org/year/2020/welcome

Conference

Conference	IEEE VIS 2020
Abbreviated title	VIS 2020
Country/Territory	United States
City	Virtuell
Period	25/10/20 → 30/10/20
Internet address	http://ieeevis.org/year/2020/welcome

Keywords

Dimensionality Reduction
High-Dimensional Data Visualization
Secure Multi-Party Computation
Secure Visualization

ASJC Scopus subject areas

Media Technology
Modelling and Simulation

Fields of Expertise

Information, Communication & Computing

Access to Document

10.1109/VAST50239.2020.00015

https://arxiv.org/abs/2007.15591

Cite this

Xia, J, Chen, T, Zhang, L, Chen, W, Chen, Y, Zhang, X, Xie, C & Schreck, T 2020, SMAP: A Joint Dimensionality Reduction Scheme for Secure Multi-Party Visualization. in Proceedings - 2020 IEEE Conference on Visual Analytics Science and Technology, VAST 2020., 9308625, pp. 107-118, IEEE VIS 2020, Virtuell, United States, 25/10/20. https://doi.org/10.1109/VAST50239.2020.00015

@inproceedings{6607619d0b734f158e2adba40584e70d,

title = "SMAP: A Joint Dimensionality Reduction Scheme for Secure Multi-Party Visualization",

abstract = "Nowadays, as data becomes increasingly complex and distributed,data analyses often involve several related datasets that are storedon different servers and probably owned by different stakeholders.While there is an emerging need to provide these stakeholders witha full picture of their data under a global context, conventionalvisual analytical methods, such as dimensionality reduction, couldexpose data privacy when multi-party datasets are fused into asingle site to build point-level relationships. In this paper, wereformulate the conventional t-SNE method from the single-sitemode into a secure distributed infrastructure. We present a securemulti-party scheme for joint t-SNE computation, which can minimize the risk of data leakage. Aggregated visualization can beoptionally employed to hide disclosure of point-level relationships.We build a prototype system based on our method, SMAP, tosupport the organization, computation, and exploration of securejoint embedding. We demonstrate the effectiveness of our approach with three case studies, one of which is based on the deployment of our system in real-world applications.",

keywords = "Dimensionality Reduction, High-Dimensional Data Visualization, Secure Multi-Party Computation, Secure Visualization",

author = "Jiazhi Xia and Tianxiang Chen and Lei Zhang and Wei Chen and Yang Chen and Xiaolong Zhang and Cong Xie and Tobias Schreck",

note = "VAST 2020: IEEE Symposium on Visual Analytics Science and Technology ; IEEE VIS 2020, VIS 2020 ; Conference date: 25-10-2020 Through 30-10-2020",

year = "2020",

month = oct,

doi = "10.1109/VAST50239.2020.00015",

language = "English",

pages = "107--118",

booktitle = "Proceedings - 2020 IEEE Conference on Visual Analytics Science and Technology, VAST 2020",

url = "http://ieeevis.org/year/2020/welcome",

}

TY - GEN

T1 - SMAP: A Joint Dimensionality Reduction Scheme for Secure Multi-Party Visualization

AU - Xia, Jiazhi

AU - Chen, Tianxiang

AU - Zhang, Lei

AU - Chen, Wei

AU - Chen, Yang

AU - Zhang, Xiaolong

AU - Xie, Cong

AU - Schreck, Tobias

N1 - VAST 2020: IEEE Symposium on Visual Analytics Science and Technology

PY - 2020/10

Y1 - 2020/10

N2 - Nowadays, as data becomes increasingly complex and distributed,data analyses often involve several related datasets that are storedon different servers and probably owned by different stakeholders.While there is an emerging need to provide these stakeholders witha full picture of their data under a global context, conventionalvisual analytical methods, such as dimensionality reduction, couldexpose data privacy when multi-party datasets are fused into asingle site to build point-level relationships. In this paper, wereformulate the conventional t-SNE method from the single-sitemode into a secure distributed infrastructure. We present a securemulti-party scheme for joint t-SNE computation, which can minimize the risk of data leakage. Aggregated visualization can beoptionally employed to hide disclosure of point-level relationships.We build a prototype system based on our method, SMAP, tosupport the organization, computation, and exploration of securejoint embedding. We demonstrate the effectiveness of our approach with three case studies, one of which is based on the deployment of our system in real-world applications.

AB - Nowadays, as data becomes increasingly complex and distributed,data analyses often involve several related datasets that are storedon different servers and probably owned by different stakeholders.While there is an emerging need to provide these stakeholders witha full picture of their data under a global context, conventionalvisual analytical methods, such as dimensionality reduction, couldexpose data privacy when multi-party datasets are fused into asingle site to build point-level relationships. In this paper, wereformulate the conventional t-SNE method from the single-sitemode into a secure distributed infrastructure. We present a securemulti-party scheme for joint t-SNE computation, which can minimize the risk of data leakage. Aggregated visualization can beoptionally employed to hide disclosure of point-level relationships.We build a prototype system based on our method, SMAP, tosupport the organization, computation, and exploration of securejoint embedding. We demonstrate the effectiveness of our approach with three case studies, one of which is based on the deployment of our system in real-world applications.

KW - Dimensionality Reduction

KW - High-Dimensional Data Visualization

KW - Secure Multi-Party Computation

KW - Secure Visualization

UR - http://www.scopus.com/inward/record.url?scp=85099795124&partnerID=8YFLogxK

U2 - 10.1109/VAST50239.2020.00015

DO - 10.1109/VAST50239.2020.00015

M3 - Conference paper

SP - 107

EP - 118

BT - Proceedings - 2020 IEEE Conference on Visual Analytics Science and Technology, VAST 2020

T2 - IEEE VIS 2020

Y2 - 25 October 2020 through 30 October 2020

ER -

SMAP: A Joint Dimensionality Reduction Scheme for Secure Multi-Party Visualization

Abstract

Conference

Keywords

ASJC Scopus subject areas

Fields of Expertise

Access to Document

Other files and links

Fingerprint

Cite this