Search

Data anonymization is one of the main techniques used in privacy preserving data publishing, and many methods have been proposed to anonymize both individual data sets and multiple data sets. In real life, a data set is rarely isolated and two data sets published by different organizations may contain records pertaining to the same individual. For example, some patients might have visited two hospitals for the same disease, and their records are independently anonymized and published by the two hospitals. Although each published data set alone might pose a small privacy risk, the combination of two data sets may severely compromise the privacy of the individuals common to both data sets. An attack on individual privacy which uses independent data sets is called a composition attack. The topic of how to anonymize data sets to prevent a composition attack using independent data releases has not been widely investigated. In this paper, we propose a new principle to protect data sets from composition attacks. We propose a hybrid algorithm, which combines sampling, perturbation and generalization to protect data privacy from composition attacks. We experimentally demonstrate that the proposed anonymization technique significantly reduces the risk of composition attacks and also preserves good data utility. Refereed/Peer-reviewed

Personal Learning Environments (PLEs) help the learners to take control of their learning. PLEs enable the learners to set their own leaning targets and manage their learning by communicating with others in the process of learning. As latest technological advancements have brought revolution in every field of life, so as in the PLEs. Modern PLEs are the integration of a number of latest technologies i.e. blogs, Wikis, RSS feeds, where content is shaped as per the individual needs and interests of the students. Focusing on these latest aspects of the PLEs, University of South Australia initiated a three year new learning platform project in 2010, called LearnOnline, which will replace the University's current online teaching environment UniSAnet. LearnOnline was launched with a vision to foster richer learning through promoting students' active involvement in their courses and involving the students in a deeper learning experience. LearnOnline is built on modular approach and consists of different components i.e. ePortfolio, Course Outline, Lecture Recording, Copyright Monitoring, Student Email, Assessment and Feedback, Virtual Classroom, Course and Teacher Evaluation. Each component is developed separately and is fully independent. This methodology is helping the incremental implementation of the LearnOnline. As soon as a component is completed, after testing, it becomes the part of LearnOnline. In this paper, the author explains the features and workings of LearnOnline in detail and also evaluates its design methodologies.

Data publication is a simple and cost-effective approach for data sharing across organizations. Data anonymization is a central technique in privacy preserving data publications. Many methods have been proposed to anonymize individual datasets and multiple datasets of the same data publisher. In real life, a dataset is rarely isolated and two datasets published by two organizations may contain the records of the same individuals. For example, patients might have visited two hospitals for follow-up or specialized treatment regarding a disease, and their records are independently anonymized and published. Although each published dataset poses a small privacy risk, the intersection of two datasets may severely compromise the privacy of the individuals. The attack using the intersection of datasets published by different organizations is called a composition attack. Some research work has been done to study methods for anonymizing data to prevent a composition attack for independent data releases where one data publisher has no knowledge of records of another data publisher. In this chapter, we discuss two exemplar methods, a randomization based and a generalization based approaches, to mitigate risks of composition attacks. In the randomization method, noise is added to the original values to make it difficult for an adversary to pinpoint an individual’s record in a published dataset. In the generalization method, a group of records according to potentially identifiable attributes are generalized to the same so that individuals are indistinguishable. We discuss and experimentally demonstrate the strengths and weaknesses of both types of methods. We also present a mixed data publication framework where a small proportion of the records are managed and published centrally and other records are managed and published locally in different organizations to reduce the risk of the composition attack and improve the overall utility of the data usc

Data anonymization has become a major technique in privacy preserving data publishing. Many methods have been proposed to anonymize one dataset and a series of datasets of a data holder. However, no method has been proposed for the anonymization scenario of multiple independent data publishing. A data holder publishes a dataset, which contains overlapping population with other datasets published by other independent data holders. No existing methods are able to protect privacy in such multiple independent data publishing. In this paper we propose a new generalization principle (ρ,α)-anonymization that effectively overcomes the privacy concerns for multiple independent data publishing. We also develop an effective algorithm to achieve the (ρ,α)-anonymization. We experimentally show that the proposed algorithm anonymizes data to satisfy the privacy requirement and preserves high quality data utility.

Anonymization is a practical approach to protect privacy in data. The major objective of privacy preserving data publishing is to protect private information in data whereas data is still useful for some intended applications, such as building classification models. In this paper, we argue that data generalization in anonymization should be determined by the classification capability of data rather than the privacy requirement. We make use of mutual information for measuring classification capability for generalization, and propose two k-anonymity algorithms to produce anonymized tables for building accurate classification models. The algorithms generalize attributes to maximize the classification capability, and then suppress values by a privacy requirement k (IACk) or distributional constraints (IACc). Experimental results show that algorithm IACk supports more accurate classification models and is faster than a benchmark utility-aware data anonymization algorithm. Refereed/Peer-reviewed

Human genomic data is a treasure that holds rich information for bioresearch. The share of human genomic data is necessary for the continuous progress of biology, medicine and health research. However, human genomic data also contains private information of individuals. Human genomic data may be maliciously used to find out the genetic tendency for a disease, and even to track descendents and relatives of the individual. In this paper, we review some techniques for protecting privacy in sharing human genomic data and discuss problems and challenges.