IEICE Technical Committee Submission System
Conference Paper's Information
Online Proceedings
[Sign in]
... (for ESS/CS/ES/ISS)
Tech. Rep. Archives
... (for ES/CS)
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2014-11-17 17:00
[Poster Presentation] Differential Privacy on Linear Regression Model of Crowdsensing
Tran Quang Khai, Kazuto Fukuchi, Jun Sakuma (Univ. of Tsukuba) IBISML2014-47
Abstract (in Japanese) (See Japanese page) 
(in English) Learning statistic models using the data collected from crowd is one of the important tasks in the crowdsensing. Crowdsensing allows statistical model building by exploiting individuals as sensors; we must pay attention to privacy concerns. In this research, we deal with the problem of privacy preservation on model learning by means of crowdsensing in the following setting. Data contributors share data with a server, then the server learns a linear regression model from the collected data and provides the model to the model user. In the context of model learning by crowdsensing, two different privacy concerns arise: preserving privacy from server (preserving privacy for data sharing) and preserving privacy from model users (preserving privacy for model publication), but until recently there is no study dealing with both. The statistic model learnt in our framework preserves privacy from server by perturbing data with normal distribution noise, and at the same time, preserves privacy from users under guarantee of differential privacy. We show that the sample complexity of the generalization loss obtained by our framework is $O(1/sqrt{n})$. From these, when the number of data contributors is large, our proposed scheme can learn almost the same model as the non-privacy model.
Keyword (in Japanese) (See Japanese page) 
(in English) differential privacy / crowdsensing / linear regression / / / / /  
Reference Info. IEICE Tech. Rep., vol. 114, no. 306, IBISML2014-47, pp. 95-102, Nov. 2014.
Paper # IBISML2014-47 
Date of Issue 2014-11-10 (IBISML) 
ISSN Print edition: ISSN 0913-5685  Online edition: ISSN 2432-6380
Copyright
and
reproduction
技術研究報告に掲載された論文の著作権はIEICEに帰属します.(許諾番号:10GA0019/12GB0052/13GB0056/17GB0034/18GB0034)
Download PDF IBISML2014-47

Conference Information
Committee IBISML  
Conference Date 2014-11-17 - 2014-11-19 
Place (in Japanese) (See Japanese page) 
Place (in English) Nagoya Univ. 
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To IBISML 
Conference Code 2014-11-IBISML 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Differential Privacy on Linear Regression Model of Crowdsensing 
Sub Title (in English)  
Keyword(1) differential privacy  
Keyword(2) crowdsensing  
Keyword(3) linear regression  
Keyword(4)  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Tran Quang Khai  
1st Author's Affiliation University of Tsukuba (Univ. of Tsukuba)
2nd Author's Name Kazuto Fukuchi  
2nd Author's Affiliation University of Tsukuba (Univ. of Tsukuba)
3rd Author's Name Jun Sakuma  
3rd Author's Affiliation University of Tsukuba (Univ. of Tsukuba)
4th Author's Name  
4th Author's Affiliation ()
5th Author's Name  
5th Author's Affiliation ()
6th Author's Name  
6th Author's Affiliation ()
7th Author's Name  
7th Author's Affiliation ()
8th Author's Name  
8th Author's Affiliation ()
9th Author's Name  
9th Author's Affiliation ()
10th Author's Name  
10th Author's Affiliation ()
11th Author's Name  
11th Author's Affiliation ()
12th Author's Name  
12th Author's Affiliation ()
13th Author's Name  
13th Author's Affiliation ()
14th Author's Name  
14th Author's Affiliation ()
15th Author's Name  
15th Author's Affiliation ()
16th Author's Name  
16th Author's Affiliation ()
17th Author's Name  
17th Author's Affiliation ()
18th Author's Name  
18th Author's Affiliation ()
19th Author's Name  
19th Author's Affiliation ()
20th Author's Name  
20th Author's Affiliation ()
Speaker
Date Time 2014-11-17 17:00:00 
Presentation Time 180 
Registration for IBISML 
Paper # IEICE-IBISML2014-47 
Volume (vol) IEICE-114 
Number (no) no.306 
Page pp.95-102 
#Pages IEICE-8 
Date of Issue IEICE-IBISML-2014-11-10 


[Return to Top Page]

[Return to IEICE Web Page]


The Institute of Electronics, Information and Communication Engineers (IEICE), Japan