Anonymous Table Research Articles

Limiting sensitive values in an anonymized table while reducing information loss via p‐proportion

AbstractThe ‐proportion model bounds the proportion of sensitive values of a sensitive attribute in each equivalence class of an anonymized database table in order to limit the ability of a user to link an individual or entity to a sensitive value in that table. Nonsensitive values are not subject to any such constraints, which reduces the amount of anonymization needed to meet the requirements of this model. This leads to less information loss in an anonymized table. Anonymization is performed using an extension of the Mondrian algorithm that incorporates categorical attributes. Known as the adapted Mondrian algorithm, it generalizes a value of a categorical attribute to a set. Existing algorithms, by comparison, replace one value of a predefined hierarchy by another. The ‐proportion model is compared against the ()‐anonymity model using both the progressive local recoding and (adapted) Mondrian algorithms. Experiments demonstrate the advantage of ‐proportion and Mondrian over ()‐anonymity and progressive local recoding in terms of reduced information loss, measured using the normalized certainty penalty, discernibility metric, and classification metric.

Relative Privacy Threats and Learning From Anonymized Data

We consider group-based anonymization schemes, a popular approach to data publishing. This approach aims at protecting privacy of the individuals involved in a dataset, by releasing an obfuscated version of the original data, where the exact correspondence between individuals and attribute values is hidden. When publishing data about individuals, one must typically balance the learner ’s utility against the risk posed by an attacker , potentially targeting individuals in the dataset. Accordingly, we propose a unified Bayesian model of group-based schemes and a related MCMC methodology to learn the population parameters from an anonymized table. This allows one to analyze the risk for any individual in the dataset to be linked to a specific sensitive value, when the attacker knows the individual’s nonsensitive attributes, beyond what is implied for the general population. We call this relative threat analysis. Finally, we illustrate the results obtained with the proposed methodology on a real-world dataset.

Clustering-based privacy preserving anonymity approach for table data sharing

Government data sharing can effectively improve the efficiency and quality of government services and enhance the ability of providing government services. However, data sharing may bring the risk of citizen privacy leakage. It is a challenging problem on improving government governance and service levels when sharing government data while guaranteed citizens’ privacy. For the diversity types and complex attributes of government data, this paper proposes a cluster-based anonymous table data sharing privacy protection method (CATDS). Firstly, preprocessing the data table. According to the correlation degree between attributes, the clustering algorithm is used to divide the data attribute column to generate multiple tables. That can reduce the data dimension and improve the algorithm execution speed. Then clustering the table data using k-medoids clustering algorithm to generate a clustering result table that initially satisfies the ķ-anonymity requirement. That can reduce the next generalization degree and improve the data availability. Finally, anonymizing the resulting clusters through generalization technique to ensure the privacy of the shared data. By comparing the CATDS with the Incognito algorithm which is a classical ķ-anonymity algorithm, it is proved that the proposed algorithm can effectively reduce the amount of information loss and improve the availability of shared table data while protecting the private information of shared table data.

Like a phoenix from the ashes – or “sack cloth and ashes”?: The reconstitution of psychoanalytic institutions in Germany since 1945 and its consequences

Here, I attempt to formulate some thoughts about the past, present, and future of psychoanalysis and its institutions in Germany. To do this, I have employed my varied experience as a supervisor and consultant to many such psychoanalytic institutes over the past several years. Themes discussed include the history of psychoanalysis in postwar Germany, the organizational structure of German psychoanalytic institutes, and their cultures in regard to group and organizational dynamics, and political and economic aspects. Finally, I add brief thoughts about the future, taking into account recent developments relating to planned changes in laws governing psychotherapy in Germany. Further, I attempt to analyze and comment on: coming to terms with the past; how to begin after the “Zero Hour”; the form of organization of psychoanalytic institutes in Germany; missing patients and missing candidates; constructive debate and hurting people’s feelings; the lack of “detoxification” and “recycling” of the poisonous remains of psychoanalytic processes; and the future of psychoanalytic institutions in Germany. I end with an example of a typical primary task used in conducting large groups in the institutes in which I worked, and include an anonymized table listing individual interventions, their duration, and frequency. These should provide an idea of my way of working, and an overview of the dimensions of the task.

Generalization Step Analysis for Privacy Preserving Data Publishing

Data publishing generate much attention over the protection of individual privacy. In this paper, we show that it is necessary to reduce the steps of generalization in order to minimize information loss in privacy preserving data publishing, but sometimes the anonymous table on basis of the method could still be attacked when an attacker can possibly determine the privacy requirement and anonymous operations by examining the published data, or its documentation, and learn the mechanism of the anonymous algorithm. We call such an attack an advanced attack. To solve the problem, the condition of attack is analyzed, and a m-threshold model is presented to decide whether the value of quasi-identifier attribute would be continuously generalized, making use of algorithm of the GSSK(Generalization Step Safe of K-anonymity) to deal with the model. Finally, computer experiments show that the GSSK algorithm can prevent the attack with little information loss.