38 results about "Data anonymity" patented technology

The invention discloses a minimum information loss control method in business data anonymity release. The method comprises the steps that firstly, nonsensitive items in a dataset are generalized to the highest level knot; secondly, privacy checking is performed on the generalized dataset, if the confidence coefficient is larger than a privacy constraint condition p, the current dataset is partially deleted and processed, when the current dataset is partially deleted and processed, privacy checking is constantly called and stopped until the current dataset is in accordance with a data privacy protection anonymity condition, the current dataset which is in accordance with the privacy protection anonymity requirements is transferred to a minimum information control mechanism, and whether grouping and refining continue to be performed or not is determined through the front and back information loss comparison. If the information loss after grouping and refining processing is small, grouping and refining processing continues to be performed, and meanwhile the above process is performed on multiple sub-branch groups formed by one group. If the information loss increases on the contrary after grouping and refining, then the current branch group is directly released.

In a method of anonymising user data by aggregation, at least one server-side device (1) receives an anonymous aggregation command from a user client device (2). The anonymous aggregation command includes a specification of a set of users and an action to be taken. A list of users who meet the specification in the anonymous aggregation command is generated. The list of users who meet the specification is validated as meeting at least one criterion for anonymous aggregation. The action in the anonymous aggregation command is triggered to be taken in respect of the validated list of users who meet the specification.

The invention discloses a multi-domain data privacy protection method for a cloud platform, and relates to the technical field of multi-domain data privacy protection. The method aims to solve the problems that an existing multi-domain data privacy protection method generally adopts static anonymity to publish data, is not flexible enough, cannot limit the data range obtained by data analysts andcannot meet the requirement for using original data. The method comprises the steps of data anonymization treatment and original data recovery treatment. The invention aims at specific requirements indata analysis and transaction processing scenes. Corresponding privacy protection strategies are adopted. For structured data tables in different fields, the effect of reducing the cloud data privacyleakage risk is achieved through a privacy protection technology taking data anonymity as a main body. Meanwhile, certain transaction processing requirements are met. The anonymized data is stored and used for data analysis, so that the risk of privacy information leakage is reduced. The data is dynamically anonymized. The data range which can be obtained by data analysts is limited. Meanwhile, the data privacy is better protected.

A system and method for secure data management is presented. The method comprises receiving a query, performing the query and obtaining answers to it, creating certain formulas representing the answers; and determining whether there is a minimum number of distinct models of these formulas conjoined with the relevant anonymity predicates along with previous answers and general background knowledge, and when that number of models does not exist, suppressing the answers. In one aspect, the method further comprises creating formulas encoding the information conveyed by the answer; and combining the formulas encoding the information with prior information.

The invention provides a graph data anonymization method and device based on a deep neural network and a storage medium. According to the method, an anonymous graph with high anonymity and high data availability can be generated for data release, and the method comprises the following steps: 1, sampling by using a random walk strategy of a node2vec algorithm to obtain a real node sequence in graphdata, and representing nodes in the node sequence as vectors in an One-hot manner; 2, constructing a learning model of graph data features based on a deep neural network, training the learning modelby taking a real node sequence as input, and optimizing model parameters to obtain a trained learning model; 3, inputting a real node sequence into the trained learning model, and outputting a randomwalk sequence of the simulated nodes; 4, adding noise meeting a differential privacy mechanism to the obtained random walk sequence of the simulated node, and then performing synthesis to obtain an anonymous graph.

The invention discloses a personalized medical image management system. The system comprises a data receiving module, a data storage module, a data anonymity module, an authority control module, a keyimage storage module, and a key image viewing module; the data receiving module is used for receiving a DICOM image file corresponding to primary image examination and associating the DICOM image file with an item to which the DICOM image file belongs; the data storage module is used for importing and storing the DICOM image file; the data anonymity module is used for anonymously processing the patient name, the patient number, the gender, the birth date and the hospital name in the DICOM image file; the authority control module is used for an administrator to set a specific film reader of the item, specific authority of each film reader, a specific quality inspector and specific authority of the specific quality inspector; the key image storage module is used for storing key image information formed after the DICOM format image is marked, measured and processed by the film reader, and the key image information comprises a key image, marking time and a marker; and the key image viewing module is used for viewing the stored key image information by the film reader and the quality inspector according to the corresponding authority.

The invention discloses a track data anonymity removing method based on deep learning, which comprises the following steps: firstly, under preset division precision, dividing a space into a pluralityof subspaces similar to a grid shape; dividing the trajectory data into two groups, one group being a known user trajectory and the other group being an anonymous user trajectory; splitting the two groups of trajectories into trajectory segments, and mapping the trajectory segments into grid data to obtain a group of grid identification sequences corresponding to the trajectories; secondly, inputting the set of grid identification sequences in the previous step into an improved DenseNet model, taking a known user track as a training set, and taking an anonymous user track as a test set; finally, classifying the tracks through a model, enabling the anonymous data to correspond to data in a known user data set, and finding out real user information corresponding to the anonymous user track data. According to the invention, the de-anonymization of the private trajectory data is realized to verify the effectiveness of the attack, so that the privacy risk of the anonymous trajectory data isrevealed.

The invention relates to a differential privacy protection method and system for a track time mode, and belongs to the field of data mining. The method comprises the following steps: firstly, based onan anonymization thought, carrying out coarse-grained disturbance on track time data by using kanonymity, and anonymously hiding data in an original single time period into a whole day; then, using aLaplace mechanism for carrying out fine-grained disturbance on a timestamp of a track; and finally, limiting the noise disturbance error within a fixed range based on a truncation Laplace mechanism,thereby improving the precision of the published data. The problem of individual privacy leakage caused by periodicity of the track time mode is solved.

The invention discloses a privacy protection method for data integration of a DaaS application, and the method comprises the following steps: 1, carrying out the multi-round cooperation among tenantsunder the condition that the anonymity of data is met, and each round employing an attribute plus a fine data set with the maximum information gain; 2, setting reputation levels of the cloud service providers, and dividing the cloud service providers according to the reputation levels; 3, for the cloud service providers with the reputation levels lower than the preset reputation level, using a privacy protection mechanism based on segmentation, hiding the incidence relation between data, ensuring value domain equilibrium distribution of attributes in a grouping equalization mode, and preventing the cloud service providers from leaking data privacy of tenants; and for the cloud service providers higher than the preset reputation level, verifying correctness and integrity of data returned bythe cloud service providers by adopting a classification index tree data structure. According to the method, through a classification index tree data structure, the cloud tenants have the capabilityof verifying the correctness and integrity of a result set returned by the cloud service provider.

The invention discloses a MapReduce-based distributed data anonymity processing method, which comprises a server side and computer terminals, wherein an original data table is stored in the server side to carry out global generalization on data and give a generalized lattice which is likely to meet k- anonymity; the server side utilizes a method of bisection to allocate a computational node to each computer terminal; each computer terminal carries out computation in parallel and returns a value to the server side according to a computation condition; if the return value does not meet k- anonymity, the server side sends a descendant node determined by the method of bisection to each computer node, otherwise the server side sends an ancestor node determined by the method of bisection to a computer; and each computer terminal recalculates according to a new node given by the server side until all nodes which meet k- anonymity are found. The method solves the trouble between explosive data growth and existing server storage and computational capabilities, and the efficiency of massive data processing is improved.