An Approach for Distributing Sensitive Values in k-Anonymity

Widodo, Widodo; Budiardjo, Eko K.; Wibowo, Wahyu Catur; Achsan, Harry Tursulistyono Yani

doi:10.1109/iwbis.2019.8935849

Cited by 8 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The well‐known “adult” table of the UCI machine learning repository is used in Experiment 1 50 . This table has been used in previous experiments 4,7,8,30‐37,39,40,45,51‐53 . Of the

15

attributes of this table, nine are used by Wang et al 30,31 For Experiment 1, these same nine are selected, namely age , workclass , education , marital‐status , occupation , race , sex , native‐country , and salary‐class , where salary‐class is the sensitive attribute.…”

Section: Resultsmentioning

confidence: 99%

“…50 This table has been used in previous experiments. 4,7,8,[30][31][32][33][34][35][36][37]39,40,45,[51][52][53] Of the 15 attributes of this table, nine are used by Wang et al 30,31 For Experiment 1, these same nine are selected, namely age, workclass, education, marital-status, occupation, race, sex, native-country, and salary-class, where salary-class is the sensitive attribute. Following Wong et al (in the case of AKPLR), sex is generalized once, occupation and race are each generalized twice, workclass, marital-status, and native-country are each generalized three times, and age and education are each generalized four times.…”

Section: Experiments 1: Adult Tablementioning

confidence: 99%

“…More recently, the concept of (

α, k

)‐anonymity has been extended by Liu et al to customize the level of privacy for specific individuals in a table 33 . Budiardjo, Wibowo, and Achsan present another model that distinguishes sensitive from nonsensitive values, albeit with the same demand that each equivalence class have at least

k

tuples, as well as in the special context of added sensitive attributes 34. Using

ℓ

‐diversity, Xiao and Li do something similar over attributes having values that are assigned one of three levels of sensitivity 35 .…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Limiting sensitive values in an anonymized table while reducing information loss via p‐proportion

Dosselmann

Hamilton

2021

Security and Privacy

View full text Add to dashboard Cite

The p‐proportion model bounds the proportion of sensitive values of a sensitive attribute in each equivalence class of an anonymized database table in order to limit the ability of a user to link an individual or entity to a sensitive value in that table. Nonsensitive values are not subject to any such constraints, which reduces the amount of anonymization needed to meet the requirements of this model. This leads to less information loss in an anonymized table. Anonymization is performed using an extension of the Mondrian algorithm that incorporates categorical attributes. Known as the adapted Mondrian algorithm, it generalizes a value of a categorical attribute to a set. Existing algorithms, by comparison, replace one value of a predefined hierarchy by another. The p‐proportion model is compared against the (α,k)‐anonymity model using both the progressive local recoding and (adapted) Mondrian algorithms. Experiments demonstrate the advantage of p‐proportion and Mondrian over (α,k)‐anonymity and progressive local recoding in terms of reduced information loss, measured using the normalized certainty penalty, discernibility metric, and classification metric.

show abstract

“…The well‐known “adult” table of the UCI machine learning repository is used in Experiment 1 50 . This table has been used in previous experiments 4,7,8,30‐37,39,40,45,51‐53 . Of the

15

Section: Resultsmentioning

confidence: 99%

Section: Experiments 1: Adult Tablementioning

confidence: 99%

“…More recently, the concept of (

α, k

k

tuples, as well as in the special context of added sensitive attributes 34. Using

ℓ

‐diversity, Xiao and Li do something similar over attributes having values that are assigned one of three levels of sensitivity 35 .…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Limiting sensitive values in an anonymized table while reducing information loss via p‐proportion

Dosselmann

Hamilton

2021

Security and Privacy

View full text Add to dashboard Cite

show abstract

“…Kai-Cheng Liu et al [7] introduced optimized data de-identification using multidimensional k-anonymity and proved that it provides more reliable anonymous data and reduce the information loss rate. Widodo et al [8] proposed an approach for distributing sensitive values in k-anonymity which outperformed systematic clustering when a high-sensitive value is distributed. Ping Zhao et al [9] proposed a non-asymptotic bound on the performance of k-anonymity against information disclosure, taking into consideration intruder's background knowledge.…”

Section: Literature Reviewmentioning

confidence: 99%

Clustering Dissimilar Tuples: A Stronger Notion of Privacy

2020

IJETER

View full text Add to dashboard Cite

Identity disclosure and attribute disclosure have always been a major concern while publishing data. k-anonymity tries to solve identity disclosure but doesn't prevent attribute disclosure which leads to homogeneity and background knowledge attack. Preserving privacy of an individual is becoming more challenging due to increasing number of homogeneity and background knowledge attacks. l-diversity model has been proposed to thwart these attacks but it doesn't fulfil its obligations. Several authors found l-diversity model to be inadequate, hence they put forth another model called t-closeness. Over the years, many investigations and experimentations conducted by various researchers shows that t-closeness does not provide a clear relationship between the threshold value t and information gain and it also shows that Earth mover's distance, a distance metric used by t-closeness model, becomes complex with multiple sensitive attributes. In view of this challenge, we propose a stronger notion of privacy called Clustering Dissimilar Tuples (CDT) to thwart homogeneity and background knowledge attack by formalizing the idea of processing the original dataset initially wherever these attacks possibly occur. These attacks are found to occur in the tuples of sensitive attributes. Hence CDT processes the tuples of sensitive attributes to form equivalence classes consisting of dissimilar tuples. Through experimental evaluations, we show that CDT is practical and can be implemented efficiently with minimum utility loss and maximum privacy gain.

show abstract

A Data Desensitization Algorithm for Privacy Protection Electric Power Industry

Tang

Zhao

Wang

et al. 2020

IOP Conf. Ser.: Mater. Sci. Eng.

View full text Add to dashboard Cite

In view of the data security problems faced by big data technology in the development of electric power industry, this paper proposes a power big data desensitization algorithm applied to privacy protection, that is, a binary K-clustering algorithm (BKC-LDA) based on K-anonymity and L diversity. First, in order to reduce the computational complexity, a classification attribute is determined to classify the data table initially, and the equivalent class number K and the sensitive attribute value category L are limited according to the number of the original ancestor in the source data table. Then, considering the influence of the change of the internal range of the attribute value on the clustering, the equation for calculating the distance between the original ancestors with weight is established, using the idea of greedy and binary K clustering to classify the data table initially sets are clustered and generalized. In addition, this big data desensitization method determines the security level policies of different permissions by adjusting the size of K and L. Our algorithm can adapt to the desensitization of power big data with different attributes in different scenarios. It can not only fully mine the value of data, but also effectively protect the privacy of users.

show abstract

An Approach for Distributing Sensitive Values in k-Anonymity

Cited by 8 publications

References 17 publications

Limiting sensitive values in an anonymized table while reducing information loss via p‐proportion

Limiting sensitive values in an anonymized table while reducing information loss via p‐proportion

Clustering Dissimilar Tuples: A Stronger Notion of Privacy

A Data Desensitization Algorithm for Privacy Protection Electric Power Industry

Contact Info

Product

Resources

About