Differential private random forest

Abhijit Patil, Sanjay Singh

Research output: Chapter in Book/Report/Conference proceedingConference contribution

12 Citations (Scopus)

Abstract

Organizations be it private or public often collect personal information about an individual who are their customers or clients. The personal information of an individual is private and sensitive which has to be secured from data mining algorithm which an adversary may apply to get access to the private information. In this paper we have consider the problem of securing these private and sensitive information when used in random forest classifier in the framework of differential privacy. We have incorporated the concept of differential privacy to the classical random forest algorithm. Experimental results shows that quality functions such as information gain, max operator and gini index gives almost equal accuracy regardless of their sensitivity towards the noise. Also the accuracy of the classical random forest and the differential private random forest is almost equal for different size of datasets. The proposed algorithm works for datasets with categorical as well as continuous attributes.

Original languageEnglish
Title of host publicationProceedings of the 2014 International Conference on Advances in Computing, Communications and Informatics, ICACCI 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages2623-2630
Number of pages8
ISBN (Electronic)9781479930791
DOIs
Publication statusPublished - 01-01-2014
Event3rd International Conference on Advances in Computing, Communications and Informatics, ICACCI 2014 - Delhi, India
Duration: 24-09-201427-09-2014

Conference

Conference3rd International Conference on Advances in Computing, Communications and Informatics, ICACCI 2014
CountryIndia
CityDelhi
Period24-09-1427-09-14

Fingerprint

Data mining
Mathematical operators
Classifiers

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Information Systems

Cite this

Patil, A., & Singh, S. (2014). Differential private random forest. In Proceedings of the 2014 International Conference on Advances in Computing, Communications and Informatics, ICACCI 2014 (pp. 2623-2630). [6968348] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICACCI.2014.6968348
Patil, Abhijit ; Singh, Sanjay. / Differential private random forest. Proceedings of the 2014 International Conference on Advances in Computing, Communications and Informatics, ICACCI 2014. Institute of Electrical and Electronics Engineers Inc., 2014. pp. 2623-2630
@inproceedings{8be15b17f0434cd3acbab8746dbe842e,
title = "Differential private random forest",
abstract = "Organizations be it private or public often collect personal information about an individual who are their customers or clients. The personal information of an individual is private and sensitive which has to be secured from data mining algorithm which an adversary may apply to get access to the private information. In this paper we have consider the problem of securing these private and sensitive information when used in random forest classifier in the framework of differential privacy. We have incorporated the concept of differential privacy to the classical random forest algorithm. Experimental results shows that quality functions such as information gain, max operator and gini index gives almost equal accuracy regardless of their sensitivity towards the noise. Also the accuracy of the classical random forest and the differential private random forest is almost equal for different size of datasets. The proposed algorithm works for datasets with categorical as well as continuous attributes.",
author = "Abhijit Patil and Sanjay Singh",
year = "2014",
month = "1",
day = "1",
doi = "10.1109/ICACCI.2014.6968348",
language = "English",
pages = "2623--2630",
booktitle = "Proceedings of the 2014 International Conference on Advances in Computing, Communications and Informatics, ICACCI 2014",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

Patil, A & Singh, S 2014, Differential private random forest. in Proceedings of the 2014 International Conference on Advances in Computing, Communications and Informatics, ICACCI 2014., 6968348, Institute of Electrical and Electronics Engineers Inc., pp. 2623-2630, 3rd International Conference on Advances in Computing, Communications and Informatics, ICACCI 2014, Delhi, India, 24-09-14. https://doi.org/10.1109/ICACCI.2014.6968348

Differential private random forest. / Patil, Abhijit; Singh, Sanjay.

Proceedings of the 2014 International Conference on Advances in Computing, Communications and Informatics, ICACCI 2014. Institute of Electrical and Electronics Engineers Inc., 2014. p. 2623-2630 6968348.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Differential private random forest

AU - Patil, Abhijit

AU - Singh, Sanjay

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Organizations be it private or public often collect personal information about an individual who are their customers or clients. The personal information of an individual is private and sensitive which has to be secured from data mining algorithm which an adversary may apply to get access to the private information. In this paper we have consider the problem of securing these private and sensitive information when used in random forest classifier in the framework of differential privacy. We have incorporated the concept of differential privacy to the classical random forest algorithm. Experimental results shows that quality functions such as information gain, max operator and gini index gives almost equal accuracy regardless of their sensitivity towards the noise. Also the accuracy of the classical random forest and the differential private random forest is almost equal for different size of datasets. The proposed algorithm works for datasets with categorical as well as continuous attributes.

AB - Organizations be it private or public often collect personal information about an individual who are their customers or clients. The personal information of an individual is private and sensitive which has to be secured from data mining algorithm which an adversary may apply to get access to the private information. In this paper we have consider the problem of securing these private and sensitive information when used in random forest classifier in the framework of differential privacy. We have incorporated the concept of differential privacy to the classical random forest algorithm. Experimental results shows that quality functions such as information gain, max operator and gini index gives almost equal accuracy regardless of their sensitivity towards the noise. Also the accuracy of the classical random forest and the differential private random forest is almost equal for different size of datasets. The proposed algorithm works for datasets with categorical as well as continuous attributes.

UR - http://www.scopus.com/inward/record.url?scp=84927642269&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84927642269&partnerID=8YFLogxK

U2 - 10.1109/ICACCI.2014.6968348

DO - 10.1109/ICACCI.2014.6968348

M3 - Conference contribution

SP - 2623

EP - 2630

BT - Proceedings of the 2014 International Conference on Advances in Computing, Communications and Informatics, ICACCI 2014

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Patil A, Singh S. Differential private random forest. In Proceedings of the 2014 International Conference on Advances in Computing, Communications and Informatics, ICACCI 2014. Institute of Electrical and Electronics Engineers Inc. 2014. p. 2623-2630. 6968348 https://doi.org/10.1109/ICACCI.2014.6968348