Privacy Preserving Data Mining Techniques Research Articles

Retraction Note: An improved privacy-preserving data mining technique using singular value decomposition with three-dimensional rotation data perturbation

Retraction Note: An improved privacy-preserving data mining technique using singular value decomposition with three-dimensional rotation data perturbation

Privacy-preserving data (stream) mining techniques and their impact on data mining accuracy: a systematic literature review

This study investigates existing input privacy-preserving data mining (PPDM) methods and privacy-preserving data stream mining methods (PPDSM), including their strengths and weaknesses. A further analysis was carried out to determine to what extent existing PPDM/PPDSM methods address the trade-off between data mining accuracy and data privacy which is a significant concern in the area. The systematic literature review was conducted using data extracted from 104 primary studies from 5 reputed databases. The scope of the study was defined using three research questions and adequate inclusion and exclusion criteria. According to the results of our study, we divided existing PPDM methods into four categories: perturbation, non-perturbation, secure multi-party computation, and combinations of PPDM methods. These methods have different strengths and weaknesses concerning the accuracy, privacy, time consumption, and more. Data stream mining must face additional challenges such as high volume, high speed, and computational complexity. The techniques proposed for PPDSM are less in number than the PPDM. We categorized PPDSM techniques into three categories (perturbation, non-perturbation, and other). Most PPDM methods can be applied to classification, followed by clustering and association rule mining. It was observed that numerous studies have identified and discussed the accuracy-privacy trade-off. However, there is a lack of studies providing solutions to the issue, especially in PPDSM.

A STUDY ON PRIVACY PRESERVING BIG DATA MINING: TECHNIQUES AND CHALLENGES

The basic goal of data mining algorithms is to extract previously undiscovered patterns from the data. When mining the data, sensitive and confidential information should be secured simultaneously to protect privacy. Due to the widespread use of information technology, enormous amounts of data are being produced at an exponential rate by several organisations, including hospitals, insurance providers, banks,e-commerce, and stock exchanges, making privacy a crucial concern in data mining. Anonymization, Perturbation, Generalization, and Cryptography are some of the privacy-preserving data mining techniques that have been proposed in the literature. In this study, we have reviewed all of these state of art techniques and presented a tabular comparison of work done by different authors as well as discussed the challenges of privacy preserving data mining.

CUNA: A privacy preserving medical records storage in cloud environment using deep encryption

Securing the medical images in cloud environment becomes mandatory because of massive image accumulation and information storage in cloud premises. Large accommodation of information in the cloud brings another risk of data security. Privacy preserving data mining techniques investigate privacy of data to be maintained before or either after the data mining technique. Lightweight technique to store the patient data securely in the cloud is focused. The proposed study is focused on designing a robust model in which medical records privacy is preserved using Customizable Unique node access (CUNA) architecture. The system design undergoes data preparation, encryption, cloud transfer, decryption and analysis modules. The input to the proposed system is patient health care data such as CT images, Medical records, Clinical treatment procedures etc. The CT images are encrypted with respect to their medical records, further transferred into the cloud by hiding the text based confidential data into the images. Inspite of normal cloud storage process, the proposed approach utilized deep cryptography technique to make authenticated storage mechanism adaptive in the cloud. CUNA is the customized commonly allocated key assigned at every node. The unique key word is encrypted, and utilized to transfer the privacy data of the patients. The data can be retrieved from the cloud, after the decryption of CUNA code. The correlation of the key is systematically achieved using Gaussian regression process algorithm (GRA). The proposed approach is compared with various state-of-art approaches and obtained the accuracy of 92% with image MSE of 0.0012%.

Verifiable Data Mining Against Malicious Adversaries in Industrial Internet of Things

With the large-scaled data generated from various interconnected machines and networks, Industrial Internet of Things (IIoT) provides unprecedented opportunities for facilitating data mining for industrial applications. The current IIoT architecture tends to adopt cloud computing for further timely mining IIoT data, however, the openness of security-critical IIoT becomes challenging in terms of unbearable privacy issues. Most existing privacy-preserving data mining (PPDM) techniques are designed to resist honest-but-curious adversaries (i.e., cloud servers and data users). Due to the complexity and openness in IIoT, PPDM is significantly difficult with the presence of malicious adversaries in IIoT who may incur incorrect learned models and inference results. To solve the aforementioned issues, we propose a framework to extend existing PPDM to guard linear regression against malicious behaviors (hereafter referred to as GuardLR). To prevent dishonest computations of cloud servers and inconsistent inputs of data users, we first design a privacy-preserving verifiable learning scheme for linear regression, which guarantees the correctness of learning. In this article, to avoid malicious clouds from returning incorrect inference results, we design a privacy-preserving prediction scheme with lightweight verification. Our formal security analysis shows that GuardLR achieves privacy, completeness, and soundness. Empirical experiments using real-world datasets also demonstrate that GuardLR has high computational efficiency and accuracy.

Privacy-Preserving Collaborative Data Collection and Analysis with Many Missing Values

Privacy-preserving data mining techniques are useful for analyzing various information, such as Internet of Things data and COVID-19-related patient data. However, collecting a large amount of sensitive personal information is a challenging task. In addition, this information may have missing values, which are not considered in the existing methods for collecting personal information while ensuring data privacy. Failure to account for missing values reduces the accuracy of the data analysis. In this paper, we propose a method for privacy-preserving data collection that considers many missing values. The patient data are anonymized and sent to a data collection server. The data collection server creates a generative model and a contingency table suitable for multi-attribute analysis based on expectation-maximization and Gaussian copula methods. Using differential privacy (the de facto standard) as a privacy metric, we conduct experiments on synthetic and real data, including COVID-19-related data. The results are 50--80\% more accurate than those of existing methods that do not consider missing values.

Privacy Preserving Parallel Clustering Based Anonymization for Big Data Using MapReduce Framework

ABSTRACT Big data refers to a massive volume of data collected from heterogeneous data sources including data collected from Internet of Things (IoT) devices. Big data analytics is playing a crucial role in extracting patterns that would benefit efficient and effective decision making. Processing this massive volume of data poses several critical issues such as scalability, security and privacy. To preserve data privacy, numerous privacy-preserving data mining and publishing techniques exist. Data anonymization utilizing data mining techniques for preserving an individual’s privacy is a promising approach to prevent the data against identity disclosure. In this paper, a Parallel Clustering based Anonymization Algorithm (PCAA) is proposed, and the results prove that the algorithm is scalable and also achieves a better tradeoff between privacy and utility. The MapReduce framework is used to parallelize the anonymization process for handling a huge volume of data. The algorithm performs well in terms of classification accuracy, F-measure, and Kullback–Leibler divergence metrics. Moreover, the big data generated from heterogeneous data sources are efficiently protected to meet the ever-growing requirements of the application.

Data Mining-Based Privacy Preservation Technique for Medical Dataset Over Horizontal Partitioned

The valuable information is extracted through data mining techniques. Recently, privacy preserving data mining techniques are widely adopted for securing and protecting the information and data. These techniques convert the original dataset into protected dataset through swapping, modification, and deletion functions. This technique works in two steps. In the first step, cloud computing considers a service platform to determine the optimum horizontal partitioning in given data. In this work, K-Means++ algorithm is implemented to determine the horizontal partitioning on the cloud platform without disclosing the cluster centers information. The second steps contain data protection and recover phases. In the second step, noise is incorporated in the database to maintain the privacy and semantic of the data. Moreover, the seed function is used for protecting the original databases. The effectiveness of the proposed technique is evaluated using several benchmark medical datasets. The results are evaluated using encryption time, execution time, accuracy, and f-measure parameters.

A SURVEY OF PRIVACY PRESERVING DATA MINING ALGORITHMS

By Abou el ela Abdou Hussein, Nermin Hamza & 2 more. A wide survey of different privacy preserving data mining (PPDM) algorithms and analyses of the representative techniques for PPDM, and points out their merits and demerits.

RETRACTED ARTICLE: An improved privacy-preserving data mining technique using singular value decomposition with three-dimensional rotation data perturbation

Recent advancements in data mining have given rise to a new channel of research, coined as privacy-preserving data mining (PPDM). PPDM technology allows us to derive useful information from vast amounts of data while protecting privacy of individual records. This paper proposed a methodology based on the machine learning algorithm called singular value decomposition (SVD) and 3D rotation data perturbation (RDP) for preserving privacy of data. Decomposition and dimensionality reduction helps to eliminate sensitive information, and perturbed matrix is generated. The original and perturbed data are classified using different classifiers, and the performance is measured in terms of accuracy rate. Accuracy is the degree of correlation between the absolute observation and the actual observations. Experimental results revealed that the proposed scheme outperforms by achieving excellent accuracy for matrices of different sizes.

Classification of Privacy Preserving Data Mining Algorithms: A Review

Nowadays, data from various sources are gathered and stored in databases. The collection of the data does not give a significant impact unless the database owner conducts certain data analysis such as using data mining techniques to the databases. Presently, the development of data mining techniques and algorithms provides significant benefits for the information extraction process in terms of the quality, accuracy, and precision results. Realizing the fact that performing data mining tasks using some available data mining algorithms may disclose sensitive information of data subject in the databases, an action to protect privacy should be taken into account by the data owner. Therefore, privacy preserving data mining (PPDM) is becoming an emerging field of study in the data mining research group. The main purpose of PPDM is to investigate the side effects of data mining methods that originate from the penetration into the privacy of individuals and organizations. In addition, it guarantees that the data miners cannot reveal any personal sensitive information contained in a database, while at the same time data utility of a sanitized database does not significantly differ from that of the original one. In this paper, we present a wide view of current PPDM techniques by classifying them based on their taxonomy techniques to differentiate the characteristics of each approach. The review of the PPDM methods is described comprehensively to provide a profound understanding of the methods along with advantages, challenges, and future development for researchers and practitioners.

Data mining privacy preserving: Research agenda

AbstractIn the modern days, the amount of the data and information is increasing along with their accessibility and availability, due to the Internet and social media. To be able to search this vast data set and to discover unknown useful data patterns and predictions, the data mining method is used. Data mining allows for unrelated data to be connected in a meaningful way, to analyze the data, and to represent the results in the form of useful data patterns and predictions that help and predict future behavior. The process of data mining can potentially violate sensitive and personal data. Individual privacy is under attack if some of the information leaks and reveals the identity of a person whose personal data were used in the data mining process. There are many privacy‐preserving data mining (PPDM) techniques and methods that have a task to preserve the privacy and sensitive data while providing accurate data mining results at the same time. PPDM techniques and methods incorporate different approaches that protect data in the process of data mining. The methodology that was used in this article is the systematic literature review and bibliometric analysis. This article identifieds the current trends, techniques, and methods that are being used in the privacy‐preserving data mining field to make a clear and concise classification of the PPDM methods and techniques with possibly identifying new methods and techniques that were not included in the previous classification, and to emphasize the future research directions.This article is categorized under: Commercial, Legal, and Ethical Issues > Security and Privacy

Privacy-Preserving Process Mining in Healthcare.

Process mining has been successfully applied in the healthcare domain and has helped to uncover various insights for improving healthcare processes. While the benefits of process mining are widely acknowledged, many people rightfully have concerns about irresponsible uses of personal data. Healthcare information systems contain highly sensitive information and healthcare regulations often require protection of data privacy. The need to comply with strict privacy requirements may result in a decreased data utility for analysis. Until recently, data privacy issues did not get much attention in the process mining community; however, several privacy-preserving data transformation techniques have been proposed in the data mining community. Many similarities between data mining and process mining exist, but there are key differences that make privacy-preserving data mining techniques unsuitable to anonymise process data (without adaptations). In this article, we analyse data privacy and utility requirements for healthcare process data and assess the suitability of privacy-preserving data transformation methods to anonymise healthcare data. We demonstrate how some of these anonymisation methods affect various process mining results using three publicly available healthcare event logs. We describe a framework for privacy-preserving process mining that can support healthcare process mining analyses. We also advocate the recording of privacy metadata to capture information about privacy-preserving transformations performed on an event log.

Comprehensive Survey on Big Data Privacy Protection

In recent years, the ever-mounting problem of Internet phishing has been threatening the secure propagation of sensitive data over the web, thereby resulting in either outright decline of data distribution or inaccurate data distribution from several data providers. Therefore, user privacy has evolved into a critical issue in various data mining operations. User privacy has turned out to be a foremost criterion for allowing the transfer of confidential information. The intense surge in storing the personal data of customers (i.e., big data) has resulted in a new research area, which is referred to as privacy-preserving data mining (PPDM). A key issue of PPDM is how to manipulate data using a specific approach to enable the development of a good data mining model on modified data, thereby meeting a specified privacy need with minimum loss of information for the intended data analysis task. The current review study aims to utilize the tasks of data mining operations without risking the security of individuals’ sensitive information, particularly at the record level. To this end, PPDM techniques are reviewed and classified using various approaches for data modification. Furthermore, a critical comparative analysis is performed for the advantages and drawbacks of PPDM techniques. This review study also elaborates on the existing challenges and unresolved issues in PPDM.

A reversible privacy-preserving clustering technique based on k-means algorithm

Data mining techniques can efficiently extract valuable knowledge from the data. However, data publishing and mining may result in a potential threat: privacy invasion. Privacy-Preserving Data Mining (PPDM) aims at effectively protecting the privacy while retaining the knowledge contained in the original data, which has received much attention in recent years. PPDM techniques often use methods such as swap, modification, and deletion to protect the original data so that no correlation exists between the original data and the resultant protected data. As a result, it cannot recover the original data from the protected data. However, in certain applications, a reliance on the original data to perform precision analysis is necessary, and consequently the storage of the original data is of great importance. In this paper, we use the concept of the k-means algorithm and propose a Reversible Privacy-Preserving k-means Clustering (kRPP) algorithm for protecting the clustering knowledge of a dataset. The experimental results show that adding noises to a cluster with a minimum total distance in the kRPP algorithm, there is no significant effect on the movement of centroids, Precision, Recall, and F1 when the noise ratio or noise offset ratio increases.

Privacy Preservation of Healthcare Data in Hybrid Cloud using a Hybrid Meta-Heuristics Based Sanitization Technique

Over the recent years, the expansion of cloud computing services enable hospitals and institutions to transit their healthcare data to the cloud, thus it provides the worldwide data access and on-demand high quality services at a cheaper rate. Despite the benefits of healthcare cloud services, the associated privacy issues are widely concerned by individuals and governments. Privacy risks rise when outsourcing personal healthcare records to cloud due to the sensitive nature of health information and the social and legal implications for its disclosure. Over the recent years, a privacy-preserving data mining (PPDM) technique has become a critical issue for the problems. Our goal is to design a privacy-preserving outsourcing framework under the hybrid cloud model. In this work we propose a Hybrid Ant Colony Optimization and Gravitational Search Algorithm (ACOGSA) to express the problem of hiding sensitive data through transaction deletion. Thus, it reduces the side effects of the hybrid cloud. Substantive experiments will be carried to compare the performance of the designed algorithm with the state-of-the-art approaches in terms of the side effects and database similarity (integrity). Over the past to sanitize the databases used for hiding sensitive information, a few heuristic approaches have been proposed. The method used for the comparison involves GA, PSO, ACO, and Firefly framework.

Privacy Preservation of Healthcare Data in Hybrid Cloud Using a Hybrid Meta-Heuristics Based Sanitization Technique

Over the recent years, the expansion of cloud computing services enable hospitals and institutions to transit their healthcare data to the cloud, thus it provides the worldwide data access and on-demand high quality services at a cheaper rate. Despite the benefits of healthcare cloud services, the associated privacy issues are widely concerned by individuals and governments. Privacy risks rise when outsourcing personal healthcare records to cloud due to the sensitive nature of health information and the social and legal implications for its disclosure. Over the recent years, a privacy-preserving data mining (PPDM) technique has become a critical issue for the problems. Our goal is to design a privacy-preserving outsourcing framework under the hybrid cloud model. In this work we propose a Hybrid Ant Colony Optimization and Gravitational Search Algorithm (ACOGSA) to express the problem of hiding sensitive data through transaction deletion. Thus, it reduces the side effects of the hybrid cloud. Substantive experiments will be carried to compare the performance of the designed algorithm with the state-of-the-art approaches in terms of the side effects and database similarity (integrity). Over the past to sanitize the databases used for hiding sensitive information, a few heuristic approaches have been proposed. The method used for the comparison involves GA, PSO, ACO, and Firefly framework.

Implementation of Client Controlled Privacy Preserving Data Model for Mining Decision Rules Using Decision Tree and Association Rules

The privacy-preserving data mining (PPDM) is one of the techniques which are used for mining data dynamically with preserving privacy of the end data owner. In this paper, a PPDM technique for generating the privacy-preserving decision rules is proposed and implemented. The key motive of presenting this privacy-preserving decision rule mining technique is to demonstrate how securely data is aggregated in the PPDM environment, how securely extract them and consumed them with the help of applications. In addition to comparing the state of art methods for mining privacy preserving decision rules for preparing the future directions of research. Therefore two different data models have used namely decision tree and association rule mining. The conducted experiments demonstrate that decision tree-based techniques are superior to the association rule mining based techniques for mining higher dimensional data with higher accuracy and low resource consumption. Therefore in the near future for extending this data model the two concepts are also introduced in this paper.

Subspace-based aggregation for enhancing utility, information measures, and cluster identification in privacy preserved data mining on high-dimensional continuous data

ABSTRACT Clustering is a data mining technique that has been effectively used in the last few decades for knowledge extraction. Privacy is a major problem while releasing data for clustering and therefore privacy-preserving data mining (PPDM) algorithms have been developed. Aggregation is a popular PPDM technique that has been used. However, in the last few years, certain applications require that data mining be performed on high-dimensional data. The present privacy preservation techniques perform aggregation in a univariate manner along each dimension. This affects the utility measures, information measures, and especially retention of original clusters. This paper proposes a new technique called as subspace-based aggregation (SBA). SBA categorizes the dimensions into dense and non-dense subspaces based on the density of points. Aggregation is performed separately for dense and non-dense subspaces. This approach helps to maximize utility measures, information measures, and retention of clusters. SBA is run on high-dimensional continuous datasets from UCI Machine Learning repository. SBA is compared with related work methods such as SINGLE, SIMPLE, MDAV, and PPPCA. SBA provides an improvement of 66% in utility, 400% in cluster identification, 5% in co-variance, and standard deviation.

A New Heuristic Anonymization Technique for Privacy Preserved Datasets Publication on Cloud Computing

Recent advancement in Information and Communication Technologies (ICT) demanded much of cloud services to sharing users’ private data. Data from various organizations are the vital information source for analysis and research. Generally, this sensitive or private data information involves medical, census, voter registration, social network, and customer services. Primary concern of cloud service providers in data publishing is to hide the sensitive information of individuals. One of the cloud services that fulfill the confidentiality concerns is Privacy Preserving Data Mining (PPDM). The PPDM service in Cloud Computing (CC) enables data publishing with minimized distortion and absolute privacy. In this method, datasets are anonymized via generalization to accomplish the privacy requirements. However, the well-known privacy preserving data mining technique called K-anonymity suffers from several limitations. To surmount those shortcomings, I propose a new heuristic anonymization framework for preserving the privacy of sensitive datasets when publishing on cloud. The advantages of K-anonymity, L-diversity and (α, k)-anonymity methods for efficient information utilization and privacy protection are emphasized. Experimental results revealed the superiority and outperformance of the developed technique than K-anonymity, L-diversity, and (α, k)-anonymity measure.