In recent years, privacy concerns about social network graph data publishing has increased due to the widespread use of such data for research purposes. This paper addresses the problem of identity disclosure risk of a node assuming that the adversary identifies one of its immediate neighbors in the published data. The related anonymity level of a graph is formulated and a mathematical model is proposed to solve the problem. The application of the method on a number of synthetic and real-world datasets confirms that the method is general and can be used in different contexts to produce superior results in terms of the utility of the anonymized graph.

In recent decades, in view of the widespread use of graph data in different applications, for instance in social networks, communications networks, etc. many researchers have investigated different Anonymization approaches for such data. Although relational data Anonymization is mature enough, graph data Anonymization is a challenging and relatively new field of research. One of the most important Anonymization models against identity disclosure risk in graph data addresses the number of links a node’ s neighbors have, in the graph. In this paper, an improved method is proposed that realizes this model using both edge addition and deletion to the original graph. The application of the method to a number of different real-world graphs confirms that the method can produce more useful graphs in terms of one of the most important characteristics in such data, i. e., the Average Path Length in the graph and graph structure will undergo less change.

Background and Objectives: Nowadays, with the rapid growth of social networks extracting valuable information from voluminous sources of social networks, alongside privacy protection and preventing the disclosure of unique data, is among the most challenging objects. In this paper, a model for maintaining privacy in big data is presented. Methods: The proposed model is implemented with Spark in-memory tool in big data in four steps. The first step is to enter the raw data from HDFS to RDDs. The second step is to determine m clusters and cluster heads. The third step is to parallelly put the produced tuples in separate RDDs. the fourth step is to release the anonymized clusters. The suggested model is based on a K-means clustering algorithm and is located in the Spark framework. also, the proposed model uses the capacities of RDD and Mlib components. Determining the optimized cluster heads in each tuple's content, considering data type, and using the formula of the suggested solution, leads to the release of data in the optimized cluster with the lowest rate of data loss and identity disclosure. Results: Using Spark framework Factors and Optimized Clusters in the K-means Algorithm in the proposed model, the algorithm implementation time in different megabyte intervals relies on multiple expiration time and purposeful elimination of clusters, data loss rates based on two-level clustering. According to the results of the simulations, while the volume of data increases, the rate of data loss decreases compared to FADS and FAST clustering algorithms, which is due to the increase of records in the proposed model. with the formula presented in the proposed model, how to determine the multiple selected attributes is reduced. According to the presented results and 2-anonomity, the value of the cost factor at k=9 will be at its lowest value of 0.20.Conclusion: The proposed model provides the right balance for high-speed process execution, minimizing data loss and minimal data disclosure. Also, the mentioned model presents a parallel algorithm for increasing the efficiency in anonymizing data streams and, simultaneously, decreasing the information loss rate.

Recent researches show that Data is one of the most valuable and important assets of organizations and businesses. Privacy in the dissemination of data is becoming increasingly challenging. Anonymity as one of the privacy strategies on one side, conceals the relationship between individuals and records in a metadata table and on the other side, preserves the usefulness of the data for subsequent analysis. Preventing information disclosure becomes difficult when the adversary possesses background knowledge. We propose an Anonymization framework to protect against background knowledge attack, identity disclosure, and feature disclosure. The Anonymization algorithm creates equivalence classes of records whose probability distributions extracted by background knowledge are similar. Our proposed algorithm satisfies k-anonymity and its extension too. The proposed anonymity algorithm tries to satisfy the privacy model while preserving the usefulness of the anonymous data. We verify the theoretical study by experimentation on two datasets. Experimental results show that our proposed algorithm outperforms the state of the art Anonymization approaches in terms of loss of information.

Healthcare providers may need to publish their operational data for consultation as well as to allow more researches.Consequently, a lot of personal specific data with high level of details are publicly available. This data may contain time series, such as ECG. Deidentification of time series is not enough to provide the requirement of privacy preservation. It is because, if a few numbers of time series are published, then appearing specific anomalies in them may reveal the sensitive information of an individual. The problem of privacy preserved time series publication is somewhat studied, but the issues of publishing the Ngrams of the time series, especially that of extracted from a small set of time series, are not considered well before.In this paper, we address this problem and define the kanonymity principle for the Ngram. The proposed schema aims to provide the k-Anonymization by repeating the rare n-grams to hide them in the crowd of frequent n-grams. We evaluate our method by using two datasets. Results of experiments show that our method can provide the requested anonymity level with low probability and entropy information loss.

Data collection and storage has been facilitated by the growth in electronic services, and has led to recording vast amounts of personal information in public and private organizations databases. These records often include sensitive personal information (such as income and diseases) and must be covered from others access. But in some cases, mining the data and extraction of knowledge from these valuable sources, creates the need for sharing them with other organizations. This would bring security challenges in user’ s privacy. The concept of privacy is described as sharing of information in a controlled way. In other words, it decides what type of personal information should be shared and which group or person can access and use it. “ Privacy preserving data publishing” is a solution to ensure secrecy of sensitive information in a data set, after publishing it in a hostile environment. This process aimed to hide sensitive information and keep published data suitable for knowledge discovery techniques. Grouping data set records is a broad approach to data Anonymization. This technique prevents access to sensitive attributes of a specific record by eliminating the distinction between a number of data set records. So far a large number of data publishing models and techniques have been proposed but their utility is of concern when a high privacy requirement is needed. The main goal of this paper to present a technique to improve the privacy and performance data publishing techniques. In this work first we review previous techniques of privacy preserving data publishing and then we present an efficient Anonymization method which its goal is to conserve accuracy of classification on anonymized data. The attack model of this work is based on an adversary inferring a sensitive value in a published data set to as high as that of an inference based on public knowledge. Our privacy model and technique uses a decision tree to prevent publishing of information that removing them provides privacy and has little effect on utility of output data. The presented idea of this paper is an extension of the work presented in [20]. Experimental results show that classifiers trained on the transformed data set achieving similar accuracy as the ones trained on the original data set.

Data repositories contain sensitive information which must be protected from unauthorized access. Existing data mining techniques can be considered as a privacy threat to sensitive data. Association rule mining is one of the utmost data mining techniques which tries to cover relationships between seemingly unrelated data in a data base.. Association rule hiding is a research area in privacy preserving data mining (PPDM) which addresses a solution for hiding sensitive rules within the data problem. Many researches have be done in this area, but most of them focus on reducing undesired side effect of deleting sensitive association rules in static databases. However, in the age of big data, we confront with dynamic data bases with new data entrance at any time. So, most of existing techniques would not be practical and must be updated in order to be appropriate for these huge volume data bases. In this paper, data Anonymization technique is used for association rule hiding, while parallelization and scalability features are also embedded in the proposed model, in order to speed up big data mining process. In this way, instead of removing some instances of an existing important association rule, generalization is used to anonymize items in appropriate level. So, if necessary, we can update important association rules based on the new data entrances. We have conducted some experiments using three datasets in order to evaluate performance of the proposed model in comparison with Max-Min2 and HSCRIL. Experimental results show that the information loss of the proposed model is less than existing researches in this area and this model can be executed in a parallel manner for less execution time.