On Using Linear Diophantine Equations for in-Parallel Hiding of Decision Tree Rules
Data sharing among organizations has become an increasingly common procedure in several areas such as advertising, marketing, electronic commerce, banking, and insurance sectors. However, any organization will most likely try to keep some patterns as hidden as possible once it shares its datasets wi...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2019-01-01
|
Series: | Entropy |
Subjects: | |
Online Access: | http://www.mdpi.com/1099-4300/21/1/66 |
id |
doaj-cc73144c9c4f4552a8e0f520f5cb3d5c |
---|---|
record_format |
Article |
spelling |
doaj-cc73144c9c4f4552a8e0f520f5cb3d5c2020-11-25T00:39:05ZengMDPI AGEntropy1099-43002019-01-012116610.3390/e21010066e21010066On Using Linear Diophantine Equations for in-Parallel Hiding of Decision Tree RulesGeorgios Feretzakis0Dimitris Kalles1Vassilios S. Verykios2School of Science and Technology, Hellenic Open University, Patras 263 35,GreeceSchool of Science and Technology, Hellenic Open University, Patras 263 35,GreeceSchool of Science and Technology, Hellenic Open University, Patras 263 35,GreeceData sharing among organizations has become an increasingly common procedure in several areas such as advertising, marketing, electronic commerce, banking, and insurance sectors. However, any organization will most likely try to keep some patterns as hidden as possible once it shares its datasets with others. This paper focuses on preserving the privacy of sensitive patterns when inducing decision trees. We adopt a record augmentation approach to hide critical classification rules in binary datasets. Such a hiding methodology is preferred over other heuristic solutions like output perturbation or cryptographic techniques, which limit the usability of the data, since the raw data itself is readily available for public use. We propose a look ahead technique using linear Diophantine equations to add the appropriate number of instances while maintaining the initial entropy of the nodes. This method can be used to hide one or more decision tree rules optimally.http://www.mdpi.com/1099-4300/21/1/66decision treesprivacy preservingDiophantine equationshiding rulesentropyinformation gaindata sharing |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Georgios Feretzakis Dimitris Kalles Vassilios S. Verykios |
spellingShingle |
Georgios Feretzakis Dimitris Kalles Vassilios S. Verykios On Using Linear Diophantine Equations for in-Parallel Hiding of Decision Tree Rules Entropy decision trees privacy preserving Diophantine equations hiding rules entropy information gain data sharing |
author_facet |
Georgios Feretzakis Dimitris Kalles Vassilios S. Verykios |
author_sort |
Georgios Feretzakis |
title |
On Using Linear Diophantine Equations for in-Parallel Hiding of Decision Tree Rules |
title_short |
On Using Linear Diophantine Equations for in-Parallel Hiding of Decision Tree Rules |
title_full |
On Using Linear Diophantine Equations for in-Parallel Hiding of Decision Tree Rules |
title_fullStr |
On Using Linear Diophantine Equations for in-Parallel Hiding of Decision Tree Rules |
title_full_unstemmed |
On Using Linear Diophantine Equations for in-Parallel Hiding of Decision Tree Rules |
title_sort |
on using linear diophantine equations for in-parallel hiding of decision tree rules |
publisher |
MDPI AG |
series |
Entropy |
issn |
1099-4300 |
publishDate |
2019-01-01 |
description |
Data sharing among organizations has become an increasingly common procedure in several areas such as advertising, marketing, electronic commerce, banking, and insurance sectors. However, any organization will most likely try to keep some patterns as hidden as possible once it shares its datasets with others. This paper focuses on preserving the privacy of sensitive patterns when inducing decision trees. We adopt a record augmentation approach to hide critical classification rules in binary datasets. Such a hiding methodology is preferred over other heuristic solutions like output perturbation or cryptographic techniques, which limit the usability of the data, since the raw data itself is readily available for public use. We propose a look ahead technique using linear Diophantine equations to add the appropriate number of instances while maintaining the initial entropy of the nodes. This method can be used to hide one or more decision tree rules optimally. |
topic |
decision trees privacy preserving Diophantine equations hiding rules entropy information gain data sharing |
url |
http://www.mdpi.com/1099-4300/21/1/66 |
work_keys_str_mv |
AT georgiosferetzakis onusinglineardiophantineequationsforinparallelhidingofdecisiontreerules AT dimitriskalles onusinglineardiophantineequationsforinparallelhidingofdecisiontreerules AT vassiliossverykios onusinglineardiophantineequationsforinparallelhidingofdecisiontreerules |
_version_ |
1725295215300837376 |