Enhancing random projection with independent and cumulative additive noise for privacy-preserving data stream mining

Denham, Benjamin; Pears, Russel; Naeem, M. Asif

doi:10.1016/j.eswa.2020.113380

Cited by 19 publications

(24 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The rotation direction can be any of the three x, y or z axes. For stream data mining, two distinguished data perturbation techniques are proposed in [10] for privacy preservation. Random projection and random translation along with two different forms of additive noise are used to develop the two perturbation techniques in [10].…”

Section: Related Workmentioning

confidence: 99%

“…For stream data mining, two distinguished data perturbation techniques are proposed in [10] for privacy preservation. Random projection and random translation along with two different forms of additive noise are used to develop the two perturbation techniques in [10]. Chamikara et al proposed a non-invertible and extendable perturbation algorithm called PABIDOT for the preservation of privacy of big data.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Enhancing the Performance of 3D Rotation Perturbation in Privacy Preserving Data Mining Using Correlation Based Feature Selection

Paul

Islam

2021

Lecture Notes on Data Engineering and Communications Technologies

View full text Add to dashboard Cite

A large amount of valuable data is being produced every day with the development of technologies. To retrieve knowledge and information from these data, mining and analysis are mandatory. But, the data may contain sensitive information of the individuals like medical diagnostic reports which they do not want to expose. Privacy preserving data mining, i.e., PPDM can help in this issue keeping the sensitive information private as well as preserving the data utility. Rotation-based perturbation technique contributes to satisfying both aspects of PPDM, i.e., individuals' privacy and data utility besides other PPDM techniques. In this work, we proposed a way for generating the triplet (set of three features) for 3D rotation perturbation technique using correlation among the features. This triplet generation is a fundamental step in 3D rotation perturbation technique. The analysis of information entropy, privacy protection and utility analysis elucidates that correlation-based triplet generation provides better data privacy and utility than existing triplet generation for 3D rotation perturbation technique.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Enhancing the Performance of 3D Rotation Perturbation in Privacy Preserving Data Mining Using Correlation Based Feature Selection

Paul

Islam

2021

Lecture Notes on Data Engineering and Communications Technologies

View full text Add to dashboard Cite

show abstract

“…Here, Artificial Neural Networks (ANN), Principal Component Analysis (PCA), Random Forests (RF), and Support Vectors machines (SVM) [ 19 , 20 ] are the latest machine learning algorithm created for the learning procedure that forms the data into two classes, such as disease affected and normal [ 21 ]. However, if the dataset has been highly partial, it cannot be utilized for decision making and data analysis [ 22 , 23 ].…”

Section: Introductionmentioning

confidence: 99%

An Efficient and Privacy-Preserving Scheme for Disease Prediction in Modern Healthcare Systems

Shynu

Chowdhary

Iwendi

et al. 2022

Sensors

View full text Add to dashboard Cite

With the Internet of Things (IoT), mobile healthcare applications can now offer a variety of dimensionalities and online services. Disease Prediction Systems (DPS) increase the speed and accuracy of diagnosis, improving the quality of healthcare services. However, privacy is garnering an increasing amount of attention these days, especially concerning personal healthcare data, which are sensitive. There are a variety of prevailing privacy preservation techniques for disease prediction that are rendered. Nonetheless, there is a chance of medical users being affected by numerous disparate diseases. Therefore, it is vital to consider multi-label instances, which might decrease the accuracy. Thus, this paper proposes an efficient privacy-preserving (PP) scheme for patient healthcare data collected from IoT devices aimed at disease prediction in the modern Health Care System (HCS). The proposed system utilizes the Log of Round value-based Elliptic Curve Cryptography (LR-ECC) to enhance the security level during data transfer after the initial authentication phase. The authorized healthcare staff can securely download the patient data on the hospital side. Utilizing the Herding Genetic Algorithm-based Deep Learning Neural Network (EHGA-DLNN) can test these data with the trained system to predict the diseases. The experimental results demonstrate that the proposed approach improves prediction accuracy, privacy, and security compared to the existing methods.

show abstract

“…Vol.65: e22210339, 2022 www.scielo.br/babt Different strategies like encryption and cryptography are adopted for covering, concealment, irritation, and anonymization of information. The authors [38] demonstrated that the gatherings scale is frequently elevated, and calculations result in significant computational and correspondence costs. A homomorphic encryption plot and a protected re-appropriated correlation conspire [12] were proposed for security, and several DNA cryptography techniques were adopted for security concerns.…”

Section: Introductionmentioning

confidence: 99%

Bit Incorporated Codon Positional Encoder (BICPE) Algorithm for Privacy Preservation in Vertically Partitioned Data in Cloud

Manithurai

Nagaraju

2022

Braz. arch. biol. technol.

View full text Add to dashboard Cite

Current days an enormous amount of data is put away and recovered from the cloud because of its benefits like stowage size, assets pooling, and multi-occupancy. Because every administration is accessible through cyberspace, there is a lot of ambiguity about security and protection at various levels. This work proposed a privacy-preserving technique by applying Deoxyribonucleic Acid (DNA) coding-based novel Bit Incorporated Codon Positional Encoder (BICPE) algorithm for vertically divided cloud information. The transactional items are scrambled with a replacement of the DNA codon. Fictitious exchanges are appended unilaterally into the private database before encryption by the data owner to alleviate frequency analysis attacks of the items. The proposed BICPE algorithm encrypts the transactional data into A, C, G, T codons to upgrade the safety of the transactional items by replacement, operation, and intricacy. The concepts of binary coding and random values are utilized in DNA sequencing to encode the data. In this method, both the sender and the receiver will share a common DNA sequence for encryption and decryption process. The performance of the proposed technique is compared with the traditional Homomorphic Encryption (HE) in terms of support threshold values (Ts) for different items in a transaction. The exploratory outcomes demonstrate that the BICPE algorithm outperforms HE as far as ciphertext size, execution time, and privacy. Hence, the newly proposed technique is more efficient and offers better performance in terms of privacy. HIGHLIGHTS• Fake transactions are inserted into database to hide item frequency.• The input binary bits are grouped into N sets with length as four for encryption.• Each item in a data set is encoded with A,C,G,T DNA codons.• Encrypted Association rules are built for frequent items in a transaction.

show abstract

Enhancing random projection with independent and cumulative additive noise for privacy-preserving data stream mining

Cited by 19 publications

References 26 publications

Enhancing the Performance of 3D Rotation Perturbation in Privacy Preserving Data Mining Using Correlation Based Feature Selection

Enhancing the Performance of 3D Rotation Perturbation in Privacy Preserving Data Mining Using Correlation Based Feature Selection

An Efficient and Privacy-Preserving Scheme for Disease Prediction in Modern Healthcare Systems

Bit Incorporated Codon Positional Encoder (BICPE) Algorithm for Privacy Preservation in Vertically Partitioned Data in Cloud

Contact Info

Product

Resources

About