Analyzing and Performing Privacy Preserving Data Mining on Medical Databases

Privacy Preserving Data Mining (PPDM) is an area that deals with data mining techniques that allow the data to be mined while keeping its privacy intact. PPDM comes into play when different parties want to involve in collaborative data mining and wish to collectively mine their data to extract knowledge while keeping their data private. The preservation of privacy of data is also of utmost importance when we want to publish any data as open data. This study includes a detailed review of some of the recent techniques proposed in this area. The biggest challenge in devising any privacy preserving data mining technique is to achieve a good balance between privacy and utility of the privacy preserved data.

With the growing technology and amount of data, the use of cloud and centralized servers is increasing. When the data of the individuals is stored by the third party it leads to the insecurity of misuse of their information in the individuals. Data Mining is carried on this data stored over cloud to get useful information, analysis and decision making purposes. But along with the process of mining, privacy preservation is the key concern. While performing data mining over large set of data there are chances of privacy violation, loss of information, loss of privacy of individuals. While concentrating only on preserving privacy can increase inefficiency and complexity of data mining techniques. Techniques used for achieving privacy lead to a significant information loss. The methods concerned about information loss generally lack behind in achieving a good level of privacy. Thus it is important to find the efficient data mining technique which can preserve privacy without causing high loss of information. This paper analyses various privacy preserving data mining techniques for their pros and cons and then gives a proposed concept to achieve privacy with minimum information loss.

A typical service provided by a Smart City is a Home Energy Management System (HEMS), an automated system that uses sensor networks to collect usage of home appliances and devices such that the collected data are accessible by the local government. For example, Smart meters, installed within individual homes, can send measures of electricity consumption for analysis in the context of the global community. Such data can be used by a forecasting tool to avoid electrical outage and improve the overall management of electricity usage by the community. However, the data collected may be protected by individual privacy laws and it is possible that the network may be violated by intruders to gain access to the lifestyle of individuals within their home. For example, an intruder may be able to identify active appliances and the number of individuals present within the residence. There is a requirement to protect the data emanating from these sensors to preserve this individual privacy information. Thus, a data concierge service capable of analyzing individual HEMS data is required; such a service would improve the lifestyle by allowing access to HEMS data by a third-party analyst. Although such a service may not require raw data, one must evaluate the trade-off between the preservation of privacy information and the effectiveness of the impact of the data services on the entire community. In this paper, we implemented a prototype application in conjunction with two noise-based perturbations. The application recommends improvements to household electricity usage based on its measured electricity consumption collected by smart meters. The perturbations provide a method to support Privacy Preserving Data Mining (PPDM) techniques. Evaluation of the performance results of the prototype applications illustrates the influence of the PPDM in improving the accuracy of said application.

Fast algorithms for hiding sensitive high-utility itemsets in privacy-preserving utility mining

The trends of data mining on healthcare data for improving medical services have increased because of the electronic healthcare record(EHR) system, which collects a massive amount of data on a daily basis. In the current scenario, hospital maintains its EHR system and stores the detailed information of patients. Data mining for healthcare improvement requires the data from all the EHR systems located at a different location to be stored at the central data mining server. Collection of healthcare data at some untrusted central data mining server raises privacy threats. Healthcare data contains patients’ private information and sharing this information for data mining creates privacy issues. Most of the previous research either focused on k-anonymity technique which causes information loss and decreases data mining accuracy or privacy preserving data mining which is focused on only specific data mining technique. We adopt source anonymous technique as privacy preserving scheme and present a novel scheme for healthcare data collection and mining in this paper. Our scheme collects data from all EHR systems without any information loss and stores at a single central data mining server, also ensuring privacy is preserved. Central data mining server helps to analyze the collected data with different data mining techniques (Association rule mining, Classification, Clustering, etc.) without the involvement of EHR systems. Our scheme is collusion resilient against central data mining server and EHR systems. Theoretical and experimental analysis show the efficiency of our scheme in terms of computation and communication cost. The experimental results using Heart disease dataset show the advantage to EHR systems using the proposed approach in terms of disease prediction accuracy.

Maintaining data mining accuracy on distorted datasets is an important issue in privacy preserving data mining. Using matrix approximation, we propose several efficient and flexible techniques to address this issue, and utilize unique characteristics of matrix factorization to maintain data pattern. We use the support vector machine classification to compare accuracy maintenance after data distortion by different methods. With better performance than some classical data perturbation approaches, nonnegative matrix factorization and singular value decomposition are considered to be promising techniques for privacy preserving data mining. Experimental results demonstrate that mining accuracy on the distorted data used these methods is almost as good as that on the original data, with added property of privacy preservation. It indicates that the matrix factorization-based data distortion schemes perturb only confidential attributes to meet privacy requirements while preserving general data pattern for knowledge extraction.

Due to advances in communication and storage technologies, there is recent exponential increase in data generated and collected. Data mining has become a necessity to enable easier and efficient means of data processing. Data mining is a process of harvesting previously unknown information from existing data and utilizing such insight for business decision making. However, in sectors such as health, exposure of sensitive information is not a trivial issue, despite benefits of data mining. Hence, researchers have proposed various approaches on seeking valuable insight from data but being cautious to protect private data and information. The purpose of this paper is to provide a conceptual review of approaches by researchers in Privacy-Preserving Data Mining (PPDM). This paper further presents a classification of PPDM techniques and provide a clear description of distinctions of one class of techniques from the other, guided by discussions in literature. Lastly, an evaluation criteria which can be used as a building block towards standardizing evaluation criteria for PPDM techniques is proposed.

Association Rule Hiding methodology is a privacy preserving data mining technique that sanitizes the original database by hide sensitive association rules generated from the transactional database. The side effect of association rules hiding technique is to hide certain rules that are not sensitive, failing to hide certain sensitive rules and generating false rules in the resulted database. This affects the privacy of the data and the utility of data mining results. In this paper, a method called Gene Patterned Association Rule Hiding (GPARH) is proposed for preserving privacy of the data and maintaining the data utility, based on data perturbation technique. Using gene selection operation, privacy linked hidden and exposed data items are mapped to the vector data items, thereby obtaining gene based data item. The performance of proposed GPARH is evaluated in terms of metrics such as number of sensitive rules generated, true positive privacy rate and execution time for selecting the sensitive rules by using Abalone and Taxi Service Trajectory datasets.

This paper establishes the foundation for the performance measurements of privacy preserving data mining techniques. The performance is measured in terms of the accuracy of data mining results and the privacy protection of sensitive data. On the accuracy side, we address the problem of previous measures and propose a new measure, named “effective sample size”, to solve this problem. We show that our new measure can be bounded without any knowledge of the data being mined, and discuss when the bound can be met. On the privacy protection side, we identify a tacit assumption made by previous measures and show that the assumption is unrealistic in many situations. To solve the problem, we introduce a game theoretic framework for the measurement of privacy.

With the rapid development of big data and data analysis technology, a large amount of data is collected and used for data mining. However, the prediction accuracy of data mining often fails to meet the needs of third-party data users, and holders may not explicitly and openly share their data. How to share data and handle data privacy issues between entities while ensuring error-free prediction becomes a challenging problem. Therefore, privacy preserving data mining(PPDM) technology was introduced. Among them, the random perturbation method is a direct and effective method to protect data privacy by modifying some sensitive attribute values. This paper proposes a new method of data mining for privacy protection, which is based on SVM machine learning algorithm for data mining and great balances the utility and security of data. JHI and WBC data sets in machine learning database were used in the experiment, and machine learning privacy protection parameters were used for security evaluation. Experiments were conducted to compare NMF and NMFSVD algorithms, and the experiments results show that RNP privacy protection data mining method not only has less error, but also can better protect the privacy of the data set.

Data is increasing drastically day by day specifically due to the usage of social networking websites like Facebook, Twitter, etc. It is very difficult for data owners to store and manipulate such a huge size of data as it incurs more cost of maintaining resources. To overcome this issue, data owners utilize cloud resources to limit the expenses. But as a third party, the cloud can be curious which may lead to the disclosure of personal data. This led researchers to use some techniques to provide privacy in cloud storage used by data owners and this is referred to as privacy-preserving data mining (PPDM). PPDM preserves the privacy of data owners in the cloud, so the private data remains private even after the mining process. This paper focuses on some of the important PPDM techniques like data distortion, encryption, etc. It brings out an extensive survey of privacy-preserving data mining techniques, their benefits and drawbacks, and put forth the open challenges for further research.

Data is increasing rapidly with the vast development of technologies. Mining is mandatory to retrieve knowledge from these data. But, the data may contain sensitive information of the individuals such as medical diagnostic reports which they don’t want to reveal. Privacy Preserving Data Mining (PPDM) helps in this regard to keep sensitive data private as well as preserving the data utility. Besides other PPDM techniques, rotation based perturbation contributes to satisfying both aspects of PPDM i.e., individuals’ privacy and data utility. Also, normalization techniques have a significant impact in the field of PPDM. In this work, the effects of three well-known normalization approaches (i.e., z-score, min-max and decimal scaling) on rotation based perturbation techniques called 2-Dimensional Rotation Transformation (2DRT) and 3-Dimensional Rotation Transformation (3DRT), are analyzed. The effects of normalization over rotation based perturbation techniques are investigated using three benchmark classifiers with five UCI data set. The empirical analysis elucidates that 3DRT provides significant privacy preservation and data utility maintenance than 2DRT for z-score normalization.

Chapter 10 - Techniques for Privacy Preserving Data Publication in the Cloud for Smart City Applications

A number of privacy preserving techniques have been proposed recently in data mining. In this paper, we provide a review of the state-of-the-art methods for privacy preserving data mining. and discuss methods for randomization, secure multipart computation, and so on. We also make a classification for the privacy preserving data mining technologies, and analyze some works in this field, such as data distortion method for achieving privacy preserving association rule mining. Detailed evaluation criteria of privacy preserving algorithm were illustrated, which include algorithm performance, data utility, and privacy protection degree. Finally, the development of privacy preserving data mining for further directions is given.

Data mining is the process of posing queries and extracting patterns, often previously unknown from large quantities of data using pattern matching or other reasoning techniques. Data mining has many applications in security including for national security as well as for cyber security. The threats to national security include attacking buildings, destroying critical infrastructures such as power grids and telecommunication systems. Data mining techniques are being investigated to find out who the suspicious people are and who is capable of carrying out terrorist activities. Cyber security is involved with protecting the computer and network systems against corruption due to Trojan horses, worms and viruses. Data mining is also being applied to provide solutions such as intrusion detection and auditing. The first part of the presentation will discuss my joint research with Prof. Latifur Khan and our students at the University of Texas at Dallas on data mining for cyber security applications For example; anomaly detection techniques could be used to detect unusual patterns and behaviors. Link analysis may be used to trace the viruses to the perpetrators. Classification may be used to group various cyber attacks and then use the profiles to detect an attack when it occurs. Prediction may be used to determine potential future attacks depending in a way on information learnt about terrorists through email and phone conversations. Data mining is also being applied for intrusion detection and auditing. Other applications include data mining for malicious code detection such as worm detection and managing firewall policies. This second part of the presentation will discuss the various types of threats to national security and describe data mining techniques for handling such threats. Threats include non real-time threats and real-time threats. We need to understand the types of threats and also gather good data to carry out mining and obtain useful results. The challenge is to reduce false positives and false negatives. The third part of the presentation will discuss some of the research challenges. We need some form of real-time data mining, that is, the results have to be generated in real-time, we also need to build models in real-time for realtime intrusion detection. Data mining is also being applied for credit card fraud detection and biometrics related applications. While some progress has been made on topics such as stream data mining, there is still a lot of work to be done here. Another challenge is to mine multimedia data including surveillance video. Finally, we need to maintain the privacy of individuals. Much research has been carried out on privacy preserving data mining. In summary, the presentation will provide an overview of data mining, the various types of threats and then discuss the applications of data mining for malicious code detection, cyber security and national security. Then we will discuss the consequences to privacy.