Proximity Test for Sensitive Categorical Attributes in Big Data

Ouazzani, Zakariae El; Bakkali, Hanan El

doi:10.1109/cloudtech.2018.8713359

“…Thus, t-closeness comes to resist against similarity attack. Based on the algorithm proposed in [28], buckets 2 and 4 of Table 4 will remain intact since they include values belonging to different categories. However, buckets 1 and 3 of Table 4 will be changed to address the problem of similarity attack since they are corresponding to "Respiratory infection" and "Brain diseases" categories respectively.…”

Section: Figure 3 the Hierarchy Of Diseasesmentioning

confidence: 99%

“…Alternatively, permutation technique modifies only the values in the data set by substituting a record with another one. Permutation technique could be applied during the anonymization process between minimum values corresponding to each consecutive buckets existing in the data set [28]. Such permutation between records will ensure that the range and the distribution of values will remain unchanged, but the link between values and individuals will not.…”

Section: Permutationmentioning

confidence: 99%

See 1 more Smart Citation

A Classification of non-Cryptographic Anonymization Techniques Ensuring Privacy in Big Data

Ouazzani

¹

,

Bakkali

²

2022

Int. j. commun. netw. inf. secur.

View full text Add to dashboard Cite

Recently, Big Data processing becomes crucial to most enterprise and government applications due to the fast growth of the collected data. However, this data often includes private personal information that arise new security and privacy concerns. Moreover, it is widely agreed that the sheer scale of big data makes many privacy preserving techniques unavailing. Therefore, in order to ensure privacy in big data, anonymization is suggested as one of the most efficient approaches. In this paper, we will provide a new detailed classification of the most used non-cryptographic anonymization techniques related to big data including generalization and randomization approaches. Besides, the paper evaluates the presented techniques through integrity, confidentiality and credibility criteria. In addition, three relevant anonymization techniques including k-anonymity, l-diversity and t-closeness are tested on an extract of a huge real data set.

show abstract

Privacy Preserving in Digital Health

Ouazzani

¹

,

Bakkali

²

,

Sadki

³

2020

Social, Legal, and Ethical Implications of IoT, Cloud, and Edge Computing Technologies

View full text Add to dashboard Cite

Recently, digital health solutions are taking advantage of recent advances in information and communication technologies. In this context, patients' health data are shared with other stakeholders. Moreover, it's now easier to collect massive health data due to the rising use of connected sensors in the health sector. However, the sensitivity of this shared healthcare data related to patients may increase the risks of privacy violation. Therefore, healthcare-related data need robust security measurements to prevent its disclosure and preserve patients' privacy. However, in order to make well-informed decisions, it is often necessary to allow more permissive security policies for healthcare organizations even without the consent of patients or against their preferences. The authors of this chapter concentrate on highlighting these challenging issues related to patient privacy and presenting some of the most significant privacy preserving approaches in the context of digital health.

show abstract

Privacy Preserving in Digital Health

Ouazzani

¹

,

Bakkali

²

,

Sadki

³

2021

Research Anthology on Privatizing and Securing Data

2

0

View full text Add to dashboard Cite

Recently, digital health solutions are taking advantage of recent advances in information and communication technologies. In this context, patients' health data are shared with other stakeholders. Moreover, it's now easier to collect massive health data due to the rising use of connected sensors in the health sector. However, the sensitivity of this shared healthcare data related to patients may increase the risks of privacy violation. Therefore, healthcare-related data need robust security measurements to prevent its disclosure and preserve patients' privacy. However, in order to make well-informed decisions, it is often necessary to allow more permissive security policies for healthcare organizations even without the consent of patients or against their preferences. The authors of this chapter concentrate on highlighting these challenging issues related to patient privacy and presenting some of the most significant privacy preserving approaches in the context of digital health.

show abstract

Proximity Test for Sensitive Categorical Attributes in Big Data

Cited by 3 publications

References 0 publications

A Classification of non-Cryptographic Anonymization Techniques Ensuring Privacy in Big Data

A Classification of non-Cryptographic Anonymization Techniques Ensuring Privacy in Big Data

Privacy Preserving in Digital Health

Privacy Preserving in Digital Health

Contact Info

Product

Resources

About