Enterprise data linkage desensitization method and device and computer equipment

Abstract

The application relates to a linkage desensitization method, device and computer equipment for enterprise data. The method comprises the following steps: acquiring enterprise data needing data desensitization, and performing encryption processing on the enterprise data to obtain an enterprise encryption graph; based on the enterprise encryption graph, a sparse risk hypergraph is constructed, and weight values of each hyperedge set of the sparse risk hypergraph are identified; based on the sparse risk hypergraph and the weight values of each hyperedge set of the sparse risk hypergraph, the data types of each sensitive field in the enterprise data and the desensitization influence categories of each sensitive field are identified, and based on the data types of each sensitive field and the desensitization influence categories of each sensitive field, field desensitization processing is respectively performed on each sensitive field to obtain desensitized enterprise data. The method can improve the desensitization efficiency of the desensitized field.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of big data security and privacy protection, in particular to a linkage desensitization method and device for enterprise data and a computer device. BACKGROUND

[0002] With the wide application of cloud computing and big data analysis, the risk of data leakage is significantly increased when the financial, medical, and government industries share data, train models, and outsource services. The demand for data desensitization technology is increasing in the data service industry. Traditional data desensitization methods are mostly based on two-dimensional relationships such as differential privacy, K-anonymity, and L-diversity, and can only statically shield single fields or field pairs, making it difficult to capture high-order correlation risks caused by field combinations. Therefore, how to effectively avoid high-order correlation risks is the current research focus.

[0003] The prior art is a method based on hypergraph structure to capture the risk of high-order correlation. However, in the current data desensitization scheme, the huge computational complexity overhead caused by multi-field, multi-dimensional high-order desensitization and the need for quantitative calculation accuracy of the correlation risk in designing the desensitization scheme are contradictory, resulting in low efficiency of data desensitization. SUMMARY

[0004] Therefore, it is necessary to provide a linkage desensitization method and device for enterprise data, a computer device, a computer readable storage medium, and a computer program product to solve the above technical problems.

[0005] In a first aspect, the present application provides a linkage desensitization method for enterprise data, comprising:

[0006] Obtaining enterprise data that needs to be desensitized, and performing encryption processing on the enterprise data to obtain an enterprise encryption graph;

[0007] Based on the enterprise encryption graph, a sparse risk hypergraph is constructed, and the weight values of each hyperedge set of the sparse risk hypergraph are identified;

[0008] Based on the sparse risk hypergraph and the weight values of each hyperedge set of the sparse risk hypergraph, the data types of each sensitive field in the enterprise data and the desensitization influence categories of each sensitive field are identified, and each sensitive field is subjected to field desensitization processing based on the data types of each sensitive field and the desensitization influence categories of each sensitive field, to obtain desensitized enterprise data.

[0009] Optionally, the encryption processing on the enterprise data to obtain an enterprise encryption graph comprises:

[0010] In the enterprise data, field data of each field after deduplication is filtered, and based on the field data of each field, a field name of each field and a field introduction of each field are identified;

[0011] The field name of each field and the field introduction of each field are respectively subjected to string concatenation processing to obtain new field data corresponding to each field, and based on the new field data corresponding to each field, an enterprise data graph is constructed;

[0012] Through a multi-encryption strategy, data encryption processing is performed on the enterprise data graph to obtain an enterprise encrypted graph.

[0013] Optionally, based on the enterprise encrypted graph, a sparse risk hypergraph is constructed, including:

[0014] Based on the enterprise encrypted graph, a sampling matrix is filtered, and a uniqueness degree of the sampling matrix is calculated to obtain a uniqueness degree proportion corresponding to each field combination of the sampling matrix;

[0015] Based on the uniqueness degree proportion corresponding to each field combination, each target field combination is filtered through a combination filtering strategy, and each target field is used as each hyperedge set;

[0016] Based on each hyperedge set, a sparse risk hypergraph is constructed.

[0017] Optionally, the weight value of each hyperedge set of the sparse risk hypergraph is identified, including:

[0018] Based on each hyperedge set, a normalized mutual information of each hyperedge set is calculated through a normalization algorithm;

[0019] Based on the normalized mutual information of each hyperedge set, a weight value of each hyperedge set is calculated through a hyperedge weight estimation algorithm.

[0020] Optionally, based on the sparse risk hypergraph and the weight value of each hyperedge set of the sparse risk hypergraph, a data type of each sensitive field in the enterprise data and a desensitization influence category of each sensitive field are identified, including:

[0021] Based on the weight value of each hyperedge set, a risk inference processing is performed on each hyperedge set to obtain risk distribution information of the sparse risk hypergraph;

[0022] Based on the risk distribution information of the sparse risk hypergraph, a sensitive field in each field of each hyperedge set is identified, and a data type to which each sensitive field of each hyperedge set belongs is identified;

[0023] determine a score value corresponding to each sensitive field and a desensitization influence category of each sensitive field based on the data type of each sensitive field.

[0024] Optionally, the data type of each sensitive field and the desensitization influence category of each sensitive field are used to respectively perform field desensitization processing on each sensitive field to obtain desensitized enterprise data, including:

[0025] Based on the desensitization influence category of each sensitive field and the data type of each sensitive field, a field desensitization cost value corresponding to each sensitive field is calculated by a field desensitization cost algorithm.

[0026] Based on the field desensitization cost value corresponding to each sensitive field, each minimum desensitization coverage subset corresponding to the enterprise data is identified in the enterprise data by a desensitization coverage algorithm.

[0027] Based on the data type of each sensitive field in each desensitization coverage subset, field desensitization processing is performed on each minimum desensitization coverage subset by a desensitization strategy corresponding to each data type to obtain desensitized enterprise data.

[0028] In a second aspect, the present application also provides a linkage desensitization device for enterprise data, including:

[0029] An acquisition module is configured to acquire enterprise data that needs to be desensitized and perform encryption processing on the enterprise data to obtain an enterprise encryption graph.

[0030] An identification module is configured to construct a sparse risk hypergraph based on the enterprise encryption graph and identify a weight value of each hyperedge set of the sparse risk hypergraph.

[0031] A desensitization module is configured to identify a data type of each sensitive field in the enterprise data and a desensitization influence category of each sensitive field based on the sparse risk hypergraph and the weight value of each hyperedge set of the sparse risk hypergraph, and perform field desensitization processing on each sensitive field based on the data type of each sensitive field and the desensitization influence category of each sensitive field to obtain desensitized enterprise data.

[0032] Optionally, the acquisition module is specifically configured to:

[0033] Each field data of each field after deduplication is screened in the enterprise data, and a field name of each field and a field introduction of each field are identified based on the field data of each field.

[0034] The field name of each field and the field introduction of each field are respectively subjected to string splicing processing to obtain new field data corresponding to each field, and an enterprise data graph is constructed based on the new field data corresponding to each field.

[0035] The enterprise data graph is subjected to data encryption processing through a multiple encryption strategy to obtain an enterprise encryption graph.

[0036] Optionally, the identification module is specifically configured to:

[0037] Based on the enterprise encryption graph, a sampling matrix is screened, and a uniqueness degree of the sampling matrix is calculated to obtain a uniqueness degree proportion corresponding to each field combination of the sampling matrix.

[0038] Based on the uniqueness degree proportion corresponding to each field combination, each target field combination is screened through a combination screening strategy, and each target field is taken as each hyperedge set.

[0039] Based on each hyperedge set, a sparse risk hypergraph is constructed.

[0040] Optionally, the identification module is specifically configured to:

[0041] Based on each hyperedge set, the normalized mutual information of each hyperedge set is calculated through a normalization algorithm.

[0042] Based on the normalized mutual information of each hyperedge set, the weight value of each hyperedge set is calculated through a hyperedge weight estimation algorithm.

[0043] Optionally, the desensitization module is specifically configured to:

[0044] Based on the weight value of each hyperedge set, the risk inference processing of each hyperedge set is performed to obtain the risk distribution information of the sparse risk hypergraph.

[0045] Based on the risk distribution information of the sparse risk hypergraph, the sensitive fields in each hyperedge set are identified, and the data types to which each sensitive field in each hyperedge set belongs are identified.

[0046] Based on the data types of each sensitive field, the score value corresponding to each sensitive field and the desensitization influence category of each sensitive field are determined.

[0047] Optionally, the desensitization module is specifically configured to:

[0048] Based on the desensitization influence category of each sensitive field and the data type of each sensitive field, the field desensitization cost value corresponding to each sensitive field is calculated through a field desensitization cost algorithm.

[0049] Based on the field desensitization cost value corresponding to each sensitive field, each minimum desensitization cover subset corresponding to the enterprise data is identified in the enterprise data through a desensitization cover algorithm.

[0050] Based on the data type of each sensitive field in each desensitization cover subset, each minimum desensitization cover subset is subjected to field desensitization processing through a desensitization strategy corresponding to each data type, to obtain desensitized enterprise data.

[0051] In a third aspect, the present application provides a computer device. The computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the steps of the method of any one of the first aspect when executing the computer program.

[0052] In a fourth aspect, the present application provides a computer readable storage medium. A computer program is stored thereon, and the computer program implements the steps of the method of any one of the first aspect when executed by a processor.

[0053] In a fifth aspect, the present application provides a computer program product. The computer program product comprises a computer program, and the computer program implements the steps of the method of any one of the first aspect when executed by a processor.

[0054] The above enterprise data linkage desensitization method, device and computer device, by obtaining enterprise data that needs to be desensitized, and performing encryption processing on the enterprise data to obtain an enterprise encryption graph; based on the enterprise encryption graph, a sparse risk hypergraph is constructed, and the weight values of each hyperedge set of the sparse risk hypergraph are identified; based on the sparse risk hypergraph and the weight values of each hyperedge set of the sparse risk hypergraph, the data types of each sensitive field in the enterprise data and the desensitization influence categories of each sensitive field are identified, and based on the data types of each sensitive field and the desensitization influence categories of each sensitive field, each sensitive field is subjected to field desensitization processing respectively to obtain desensitized enterprise data. The scheme abstracts a large number of sensitive fields and their combinations into a hypergraph structure, and simultaneously uses the unique degree of fast calculation as a screening benchmark to make the hypergraph structure sufficiently sparse, so as to fully exhibit the potential risks brought by multi-field linkage under the conditions of limited calculation complexity and storage space, and break through the limitations of traditional single field or field pair. Then, based on the risk weight predicted by the hypergraph, the field data cost is combined to intelligently generate a minimum desensitization field set through a weighted vertex cover algorithm, to ensure that the data usability is maximized within the given residual risk upper limit, thereby providing a multi-field sparse data desensitization solution with high precision, high efficiency and high usability landing potential, and comprehensively improving the desensitization efficiency of the desensitization field. BRIEF DESCRIPTION OF DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed to be used in the embodiments or related art description will be briefly introduced. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0056] Figure 1 A flowchart of an embodiment of the linkage desensitization method of enterprise data;

[0057] Figure 2 A flowchart of an embodiment of the linkage desensitization method of enterprise data;

[0058] Figure 3 A block diagram of an embodiment of the linkage desensitization device of enterprise data;

[0059] Figure 4 An internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION

[0060] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0061] The linkage desensitization method of enterprise data provided by the embodiments of the present application can be applied to a terminal, a server, or a system including a terminal and a server, and is realized through the interaction of the terminal and the server. The terminal can be, but is not limited to, various personal computers, notebook computers, medium-sized computers, etc. The terminal abstracts a large number of sensitive fields and their combinations into a hypergraph structure, and uses the unique degree of fast calculation as a screening benchmark to make the hypergraph structure sparse enough, so that the potential risks brought by multi-field linkage can be fully exhibited in the case of limited calculation complexity and storage space, and the limitations of traditional single field or field pair are broken through. Then, based on the risk weight predicted by the hypergraph, the minimum desensitization field set is intelligently generated by the weighted vertex covering algorithm combined with the field data cost, to ensure the maximum retention of data availability within the given residual risk upper limit, thereby providing a multi-field linkage desensitization solution with high precision, high efficiency and high availability landing potential, and comprehensively improving the desensitization efficiency of the desensitization field.

[0062] In an exemplary embodiment, as shown in Figure 1 A linkage desensitization method of enterprise data is provided, and the method is applied to a terminal as an example for illustration, including the following steps S101 to S103. Among them:

[0063] Step S101, obtain enterprise data that needs to be desensitized, and encrypt the enterprise data to obtain an enterprise encryption graph.

[0064] In this embodiment, the terminal obtains enterprise data that needs to be desensitized by the enterprise in response to the information uploading operation of the system. The enterprise data can be, but is not limited to, chart data, text data, and other data contents. Before the enterprise data is obtained, the personnel associated with each relationship table or knowledge graph of the enterprise data need to be uploaded. Then, the terminal encrypts the enterprise data to obtain an enterprise encryption graph. The encryption method can be, but is not limited to, homomorphic encryption or multi-party encryption operation. The specific encryption process will be described in detail later.

[0065] Step S102, based on the enterprise encryption graph, construct a sparse risk hypergraph, and identify the weight values of each hyperedge set of the sparse risk hypergraph.

[0066] In this embodiment, the terminal constructs a sparse risk hypergraph based on the enterprise encryption graph and identifies the weight values of each hyperedge set of the sparse risk hypergraph. The sparse risk hypergraph is constructed by arranging the enterprise data with a higher risk degree in the enterprise data according to the data arrangement mode to generate the sparse risk hypergraph. Then, the terminal calculates the weight values of each hyperedge set of the sparse risk hypergraph by a normalized information calculation method. The specific calculation process will be described in detail later.

[0067] Step S103, based on the sparse risk hypergraph and the weight values of each hyperedge set of the sparse risk hypergraph, identify the data type of each sensitive field in the enterprise data and the desensitization impact category of each sensitive field, and based on the data type of each sensitive field and the desensitization impact category of each sensitive field, respectively perform field desensitization processing on each sensitive field to obtain desensitized enterprise data.

[0068] In this embodiment, the terminal identifies the data type of each sensitive field in the enterprise data and the desensitization impact category of each sensitive field based on the sparse risk hypergraph and the weight values of each hyperedge set of the sparse risk hypergraph, and based on the data type of each sensitive field and the desensitization impact category of each sensitive field, respectively performs field desensitization processing on each sensitive field to obtain desensitized enterprise data. The field desensitization method depends on the data type corresponding to different sensitive fields, that is, the desensitization scheme corresponding to sensitive fields of different data types is different. The desensitization scheme includes but is not limited to pseudonymization, data masking, or local suppression scheme. As shown in Table 1, the relationship table corresponding to the data type of each sensitive field and the desensitization impact category is:

[0069] Table 1: Corresponding relationship table of field type

[0070]

[0071] Based on the above scheme, by abstracting the mass sensitive fields and their combinations into a hypergraph structure, and using the unique degree of fast calculation as a screening criterion to make the hypergraph structure sparse enough, the potential risks brought by the multi-field linkage can be fully exhibited under the condition of limited computational complexity and storage space, breaking through the limitation of traditional single field or field pair. Then, based on the risk weight predicted by the hypergraph, the scheme combines the field data cost to intelligently generate the minimum desensitization field set through the weighted vertex covering algorithm, ensuring the maximum retention of data availability within the given residual risk limit, thereby providing a multi-field desensitization solution with high precision, high efficiency and high availability landing potential, and comprehensively improving the desensitization efficiency of the desensitization field.

[0072] Optionally, the enterprise data is encrypted to obtain an enterprise encryption graph, including: screening the field data of each field after deduplication in the enterprise data, and identifying the field name of each field and the field introduction of each field based on the field data of each field; performing string splicing processing on the field name of each field and the field introduction of each field respectively to obtain new field data corresponding to each field, and constructing an enterprise data graph based on the new field data corresponding to each field; and performing data encryption processing on the enterprise data graph through a multiple encryption strategy to obtain the enterprise encryption graph.

[0073] In this embodiment, the terminal screens the field data of each field after deduplication in the enterprise data, and identifies the field name of each field and the field introduction of each field based on the field data of each field. Specifically, first, all fields are extracted from multiple relationship tables or existing knowledge graphs, and fields related to enterprise privacy and enterprise secrets or fields that may be combined for desensitization are pre-screened through a local large model, and the DeepSeek-R1-14B large model deployed locally is used to perform semantic deduplication on the above fields from multiple tables to obtain a set of deduplicated fields , wherein M is the total number of deduplicated fields. In the field screening, the information density of each field is counted, and fields with a base of 1 or a missing rate of more than 95% are statically removed to ensure that each has statistical value.

[0074] Then, the terminal performs string splicing processing on the field name of each field and the field introduction of each field respectively to obtain new field data corresponding to each field, and constructs an enterprise data graph based on the new field data corresponding to each field. Specifically, the terminal splices the field name and the field introduction of each field as a string, and uses a pre-trained language model deployed offline locally to embed semantic information into a vector .

[0075] Finally, the terminal uses multiple encryption strategies to encrypt the target enterprise data map, obtaining an encrypted enterprise map. Specifically, the terminal encrypts the data map by... Perform homomorphic encryption or multi-party encryption operations to enable It can no longer be inversely solved by the embedded model. Here, V is stored as a row vector in the metadata table, with each row containing a field embedding vector. Data types Cardinal number Missing rate ,get .

[0076] Based on the above solution, by preprocessing and encrypting enterprise data, the risk of data leaving the domain is avoided, and the data leakage in the subsequent process is prevented from being reverse-engineered by external models, thus achieving differential privacy protection for private domain data.

[0077] Optionally, based on the enterprise encrypted graph, a sparse risk hypergraph is constructed, including: based on the enterprise encrypted graph, a sampling matrix is ​​selected, and the uniqueness of the sampling matrix is ​​calculated to obtain the uniqueness ratio corresponding to each field combination of the sampling matrix; based on the uniqueness ratio corresponding to each field combination, each target field combination is selected through a combination selection strategy, and each target field is used as a set of hyperedges; based on each set of hyperedges, a sparse risk hypergraph is constructed.

[0078] In this embodiment, the terminal filters the sampling matrix based on the enterprise encrypted graph and calculates the uniqueness of the sampling matrix to obtain the uniqueness ratio corresponding to each field combination in the sampling matrix. Then, based on the uniqueness ratio corresponding to each field combination, the terminal filters each target field combination through a combination filtering strategy and uses each target field as a hyperedge set. Finally, the terminal constructs a sparse risk hypergraph based on each hyperedge set.

[0079] Specifically, in order to construct a sparse risk hypergraph, the terminal extracts data from the M field based on the fields of each enterprise's data. Okay, we get the sampling matrix. .

[0080] Then, based on the above sampling matrix, the system can further calculate the uniqueness of the second- and third-level field combinations, as expressed in the following expression:

[0081]

[0082] Among them, this field combination It is a certain order of field combinations, ,and The value tuple representing the field combination E. The meaning is: for the field combination E, there are rows (of enterprises) that can be uniquely determined by the value of the field combination E.

[0083] Among them, the unique degree ratio will be used for preliminary screening. The terminal will calculate the of each field combination before establishing the risk hypergraph, and only keep field combinations in the hypergraph as hyperedges. Among them, the , so that the hyperedges on the hypergraph are greatly sparse, greatly reducing the calculation complexity, and thus obtaining the hyperedge set of the hypergraph, which can be written as . After that, the weight of the hyperedge E represents the re-identification risk degree of the hyperedge.

[0084] Based on the above scheme, by constructing a sparse risk hypergraph, a large number of sensitive fields and their combinations are abstracted into a hypergraph structure, and the fast calculation of the unique degree is used as a screening benchmark to make the hypergraph structure sparse enough., so as to fully display the potential risks brought by the multi-field linkage under the condition of limited calculation complexity and storage space, and break through the limitation of traditional single field or field pair.

[0085] Optionally, the weight value of each hyperedge set of the sparse risk hypergraph includes: based on each hyperedge set, calculating the normalized mutual information of each hyperedge set through a normalization algorithm; based on the normalized mutual information of each hyperedge set, calculating the weight value of each hyperedge set through a hyperedge weight estimation algorithm.

[0086] In this embodiment, the terminal calculates the normalized mutual information of each hyperedge set based on each hyperedge set through a normalization algorithm. Specifically, the normalized mutual information is a reliable weight evaluation benchmark, and its expression is:

[0087]

[0088] In order to reduce the calculation amount, the terminal approximates the calculation of the normalized mutual information Initialize all hyperedge weights in the hypergraph, and calculate the accurate normalized mutual information of 5% of the hyperedges randomly sampled therein using formula (2) as the labeled data set, these data constitute the training set and test set of HyperGNN.

[0089] Then, the terminal calculates the weight value of each hyperedge set based on the normalized mutual information of each hyperedge set through a hyperedge weight estimation algorithm. Specifically, in the training, the features of the nodes are calculated by ​Given, and do not set the output head on the node, but only set the output head on the hyperedge E. Use two layers of HyperSAGE as the backbone network to calculate the feature representation of each hyperedge:

[0090]

[0091] And use an MLP as the output head of HyperGNN to get the weight estimate of the hyperedge:

[0092]

[0093] Based on the above scheme, by combining the normalization algorithm, the mutual information of each hyperedge set is calculated, so as to obtain the weight value of each hyperedge set, and the accuracy of risk analysis of each hyperedge set is improved.

[0094] Optionally, based on the sparse risk hypergraph and the weight value of each hyperedge set of the sparse risk hypergraph, the data type of each sensitive field in the enterprise data and the desensitization influence category of each sensitive field are identified, including: based on the weight value of each hyperedge set, the risk inference processing of each hyperedge set is performed to obtain the risk distribution information of the sparse risk hypergraph; based on the risk distribution information of the sparse risk hypergraph, the sensitive field in each field of each hyperedge set is identified, and the data type to which each sensitive field in each hyperedge set belongs is identified; based on the data type of each sensitive field, the score value corresponding to each sensitive field and the desensitization influence category of each sensitive field are determined.

[0095] In this embodiment, the terminal performs risk inference processing on each hyperedge set based on the weight value of each hyperedge set to obtain the risk distribution information of the sparse risk hypergraph; specifically, the terminal considers the loss function , and the HyperGNN and MLP head are updated by back propagation. After training, a forward inference is performed on all hyperedges to obtain the final risk distribution .

[0096] Then, the terminal identifies the sensitive field in each field of each hyperedge set based on the risk distribution information of the sparse risk hypergraph, and identifies the data type to which each sensitive field in each hyperedge set belongs; based on the data type of each sensitive field, the score value corresponding to each sensitive field and the desensitization influence category of each sensitive field are determined. Among them, the risk distribution information includes the risk value corresponding to each field, then the terminal selects the field corresponding to the risk value lower than the preset risk threshold of the terminal as the sensitive field, and then performs type adaptation through Table 1 to obtain the score value corresponding to each sensitive field and the desensitization influence category of each sensitive field.

[0097] Based on the above scheme, by applying a specially designed sparse risk hypergraph, the network can be trained with 5% label data calculated based on mutual information, and can automatically capture complex interaction patterns of field combinations in the hypergraph, and generate an interpretable risk score for each high-order risk association, improving the accuracy and flexibility of desensitization decisions.

[0098] Optionally, based on the data types of the sensitive fields and the desensitization influence categories of the sensitive fields, the terminal respectively performs field desensitization processing on each sensitive field to obtain desensitized enterprise data, including: based on the desensitization influence categories of the sensitive fields and the data types of the sensitive fields, calculating the field desensitization cost values corresponding to the sensitive fields through a field desensitization cost algorithm; based on the field desensitization cost values corresponding to the sensitive fields, identifying the minimum desensitization coverage subsets corresponding to the enterprise data in the enterprise data through a desensitization coverage algorithm; based on the data types of the sensitive fields in each desensitization coverage subset, performing field desensitization processing on each minimum desensitization coverage subset through the desensitization strategies corresponding to each data type, to obtain desensitized enterprise data.

[0099] In this embodiment, the terminal calculates the field desensitization cost values corresponding to the sensitive fields based on the desensitization influence categories of the sensitive fields and the data types of the sensitive fields through a field desensitization cost algorithm. Specifically, the field desensitization cost algorithm is:

[0100]

[0101] wherein, is the desensitization field, is the field desensitization cost value of the desensitization field, is the score value, is the data type.

[0102] Then, the terminal identifies the minimum desensitization coverage subsets corresponding to the enterprise data in the enterprise data based on the field desensitization cost values corresponding to the sensitive fields through a desensitization coverage algorithm. The screening method of each minimum desensitization coverage subset is:

[0103]

[0104] wherein, is the minimum desensitization coverage subset, is the field set in the enterprise data.

[0105] Finally, the terminal performs field desensitization processing on each minimum desensitization coverage subset based on the data types of the sensitive fields in each desensitization coverage subset through the desensitization strategies corresponding to each data type, to obtain desensitized enterprise data. The terminal predefines the correspondence between different data types and desensitization strategies. Then, the terminal adapts the desensitization strategies corresponding to each data type through the above correspondence.

[0106] Based on the above scheme, by combining different data types and adapting each desensitization strategy, the accuracy and desensitization effect of data desensitization are improved. Secondly, the scheme estimates the desensitization cost by comprehensively considering the field use value and risk intensity, and automatically selects the desensitization target by using the greedy algorithm to solve the weighted vertex cover, thereby maximizing the availability of business data while ensuring that the overall desensitization risk is controllable.

[0107] The application also provides an example of linkage desensitization of enterprise data, as shown in Figure 2 The specific processing process includes the following steps:

[0108] Step S201, obtaining enterprise data that needs to be desensitized.

[0109] Step S202, filtering the field data of each field after deduplication in the enterprise data, and identifying the field name of each field and the field introduction of each field based on the field data of each field.

[0110] Step S203, respectively performing string concatenation processing on the field name of each field and the field introduction of each field, obtaining new field data corresponding to each field, and constructing an enterprise data graph based on the new field data corresponding to each field.

[0111] Step S204, performing data encryption processing on the enterprise data graph through a multiple encryption strategy to obtain an enterprise encryption graph.

[0112] Step S205, filtering a sampling matrix based on the enterprise encryption graph, and performing uniqueness calculation on the sampling matrix to obtain the uniqueness proportion corresponding to each field combination of the sampling matrix.

[0113] Step S206, based on the uniqueness proportion corresponding to each field combination, filtering each target field combination through a combination filtering strategy, and taking each target field as a hyperedge set.

[0114] Step S207, constructing a sparse risk hypergraph based on each hyperedge set.

[0115] Step S208, based on each hyperedge set, calculating the normalized mutual information of each hyperedge set through a normalization algorithm.

[0116] Step S209, based on the normalized mutual information of each hyperedge set, calculating the weight value of each hyperedge set through a hyperedge weight estimation algorithm.

[0117] Step S210, based on the weight value of each hyperedge set, performing risk inference processing on each hyperedge set to obtain risk distribution information of the sparse risk hypergraph.

[0118] Step S211, based on the risk distribution information of the sparse risk hypergraph, identifying sensitive fields in each field of each hypercode set, and identifying the data type to which each sensitive field of each hypercode set belongs.

[0119] Step S212, based on the data type of each sensitive field, determining the score value corresponding to each sensitive field and the desensitization influence category of each sensitive field.

[0120] Step S213, based on the desensitization influence category of each sensitive field and the data type of each sensitive field, calculating the field desensitization cost value corresponding to each sensitive field through the field desensitization cost algorithm.

[0121] Step S214, based on the field desensitization cost value corresponding to each sensitive field, identifying each minimum desensitization coverage subset corresponding to the enterprise data in the enterprise data through the desensitization coverage algorithm.

[0122] Step S215, based on the data type of each sensitive field in each desensitization coverage subset, performing field desensitization processing on each minimum desensitization coverage subset through the desensitization strategy corresponding to each data type to obtain desensitized enterprise data.

[0123] It should be understood that although each step in the flowchart involved in each of the above embodiments is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise stated herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, at least part of the steps in the flowchart involved in each of the above embodiments can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or steps or stages in other steps.

[0124] Based on the same inventive concept, the embodiments of the present application also provide an enterprise data linkage desensitization device for implementing the above-mentioned enterprise data linkage desensitization method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more enterprise data linkage desensitization device embodiments provided below can refer to the limitations of the enterprise data linkage desensitization method in the above text, which will not be repeated here.

[0125] In one exemplary embodiment, as shown in Figure 3 An enterprise data linkage desensitization device is provided, comprising: an acquisition module 310, an identification module 320, and a desensitization module 330, wherein:

[0126] The acquisition module 310 is configured to acquire enterprise data that needs to be desensitized, and perform encryption processing on the enterprise data to obtain an enterprise encryption graph.

[0127] The identification module 320 is configured to construct a sparse risk hypergraph based on the enterprise encryption graph, and identify weight values of each hyperedge set of the sparse risk hypergraph.

[0128] The desensitization module 330 is configured to identify data types of each sensitive field in the enterprise data and desensitization influence categories of each sensitive field based on the sparse risk hypergraph and the weight values of each hyperedge set of the sparse risk hypergraph, and perform field desensitization processing on each sensitive field based on the data types of each sensitive field and the desensitization influence categories of each sensitive field, to obtain desensitized enterprise data.

[0129] Optionally, the acquisition module 310 is specifically configured to:

[0130] In the enterprise data, field data of each field after deduplication is filtered, and a field name of each field and a field introduction of each field are identified based on the field data of each field.

[0131] String concatenation processing is performed on the field name of each field and the field introduction of each field, to obtain new field data corresponding to each field, and an enterprise data graph is constructed based on the new field data corresponding to each field.

[0132] Data encryption processing is performed on the enterprise data graph through a multiple encryption strategy, to obtain an enterprise encryption graph.

[0133] Optionally, the identification module 320 is specifically configured to:

[0134] Based on the enterprise encryption graph, a sampling matrix is filtered, and unique degree calculation is performed on the sampling matrix to obtain a unique degree proportion corresponding to each field combination of the sampling matrix.

[0135] Based on the unique degree proportion corresponding to each field combination, each target field combination is filtered through a combination filtering strategy, and each target field is taken as each hyperedge set.

[0136] Based on each hyperedge set, a sparse risk hypergraph is constructed.

[0137] Optionally, the identification module 320 is specifically configured to:

[0138] Based on each hyperedge set, normalized mutual information of each hyperedge set is calculated through a normalization algorithm.

[0139] Based on the normalized mutual information of each hyperedge set, the weight values of each hyperedge set are calculated by a hyperedge weight estimation algorithm.

[0140] Optionally, the desensitization module 330 is specifically used for:

[0141] Based on the weight values of each hyperedge set, risk inference processing is performed on each hyperedge set to obtain risk distribution information of the sparse risk hypergraph.

[0142] Based on the risk distribution information of the sparse risk hypergraph, sensitive fields in each field of each hyperedge set are identified, and the data types to which each sensitive field of each hyperedge set belongs are identified.

[0143] Based on the data types of each sensitive field, score values corresponding to each sensitive field and desensitization influence categories of each sensitive field are determined.

[0144] Optionally, the desensitization module 330 is specifically used for:

[0145] Based on the desensitization influence categories of each sensitive field and the data types of each sensitive field, field desensitization cost values corresponding to each sensitive field are calculated by a field desensitization cost algorithm.

[0146] Based on the field desensitization cost values corresponding to each sensitive field, each minimum desensitization coverage subset corresponding to the enterprise data is identified in the enterprise data by a desensitization coverage algorithm.

[0147] Based on the data types of each sensitive field in each desensitization coverage subset, field desensitization processing is performed on each minimum desensitization coverage subset by a desensitization strategy corresponding to each data type to obtain desensitized enterprise data.

[0148] Each module in the linkage desensitization device for enterprise data can be realized by software, hardware, and combinations thereof, in whole or in part. Each module described above can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a memory in a computer device in software form, so as to be called and executed by a processor to perform operations corresponding to each module.

[0149] In one exemplary embodiment, a computer device is provided, which can be a terminal, and an internal structure diagram thereof can be as shown in Figure 4As shown in the figure. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through the system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capability. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with the external terminal in a wired or wireless manner. The wireless manner can be realized through WIFI, mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by the processor to realize a surface defect detection method of an ultra-precision optical element. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0150] Those skilled in the art can understand that, Figure 4 The skilled in the art can understand that,

[0151] In one exemplary embodiment, a computer device is provided, comprising a memory and a processor, the memory stores a computer program, and the processor executes the steps of the method of any one of the first aspect when executing the computer program.

[0152] In one embodiment, a computer readable storage medium is provided, which stores a computer program, and the processor executes the steps of the method of any one of the first aspect when executing the computer program.

[0153] In one embodiment, a computer program product is provided, comprising a computer program, and the processor executes the steps of the method of any one of the first aspect when executing the computer program.

[0154] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0155] It can be understood by those skilled in the art that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, it can include the processes of the above-mentioned embodiments of each method. Any reference to memory, database or other medium used in the embodiments provided by the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (Read-Only Memory, ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric memory (Ferroelectric Random Access Memory, FRAM), phase change memory (Phase Change Memory, PCM), graphene memory, etc. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The database involved in the embodiments provided by the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided by the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0156] The technical features of the above embodiments can be combined in any way. In order to make the description simple, not all possible combinations of the technical features in the above embodiments are described, but as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0157] The above-described embodiments are merely illustrative of several embodiments of the present application, and the description is relatively specific and detailed, but should not be understood as a limitation on the scope of the patent. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of protection of the present application should be subject to the appended claims.

Claims

1. A linkage desensitization method of enterprise data, characterized in that, The method comprises: obtaining enterprise data that needs to be desensitized; screening the field data of each field after deduplication in the enterprise data, and identifying the field name of each field and the field introduction of each field based on the field data of each field; respectively performing string concatenation processing on the field name of each field and the field introduction of each field to obtain new field data corresponding to each field, and constructing an enterprise data graph based on the new field data corresponding to each field; performing data encryption processing on the enterprise data graph through a multiple encryption strategy to obtain an enterprise encrypted graph; based on the enterprise encrypted graph, screening a sampling matrix, and performing uniqueness calculation on the sampling matrix to obtain the uniqueness proportion corresponding to each field combination of the sampling matrix; based on the uniqueness proportion corresponding to each field combination, screening each target field combination through a combination screening strategy, and taking each target field as each hyperedge set; based on each hyperedge set, constructing a sparse risk hypergraph; identifying the weight value of each hyperedge set of the sparse risk hypergraph; based on the sparse risk hypergraph and the weight value of each hyperedge set of the sparse risk hypergraph, identifying the data type of each sensitive field in the enterprise data and the desensitization influence category of each sensitive field; based on the desensitization influence category of each sensitive field and the data type of each sensitive field, calculating the field desensitization cost value corresponding to each sensitive field through a field desensitization cost algorithm; based on the field desensitization cost value corresponding to each sensitive field, identifying each minimum desensitization coverage subset corresponding to the enterprise data in the enterprise data through a desensitization coverage algorithm; based on the data type of each sensitive field in each desensitization coverage subset, performing field desensitization processing on each minimum desensitization coverage subset through a desensitization strategy corresponding to each data type to obtain desensitized enterprise data.

2. The method of claim 1, wherein, The identification of the weight value of each hyperedge set of the sparse risk hypergraph comprises: based on each hyperedge set, calculating the normalized mutual information of each hyperedge set through a normalization algorithm; based on the normalized mutual information of each hyperedge set, calculating the weight value of each hyperedge set through a hyperedge weight estimation algorithm.

3. The method of claim 1, wherein, The identification of the data type of each sensitive field in the enterprise data and the desensitization influence category of each sensitive field based on the sparse risk hypergraph and the weight value of each hyperedge set of the sparse risk hypergraph comprises: based on the weight value of each hyperedge set, performing risk inference processing on each hyperedge set to obtain risk distribution information of the sparse risk hypergraph; based on the risk distribution information of the sparse risk hypergraph, identifying the sensitive field in each field of each hyperedge set and identifying the data type to which each sensitive field of each hyperedge set belongs; based on the data type of each sensitive field, determining the score value corresponding to each sensitive field and the desensitization influence category of each sensitive field.

4. A linkage desensitization device for enterprise data, characterized by, The device comprises: The acquisition module is configured to acquire enterprise data that needs to be desensitized; in the enterprise data, field data of each field after deduplication is screened, and based on the field data of each field, a field name of each field and a field introduction of each field are identified; the field name of each field and the field introduction of each field are subjected to string concatenation processing respectively to obtain new field data corresponding to each field, and based on the new field data corresponding to each field, an enterprise data graph is constructed; and the enterprise data graph is subjected to data encryption processing through a multiple encryption strategy to obtain an enterprise encryption graph. The identification module is configured to screen a sampling matrix based on the enterprise encryption graph, and perform uniqueness calculation on the sampling matrix to obtain a uniqueness proportion corresponding to each field combination of the sampling matrix; based on the uniqueness proportion corresponding to each field combination, each target field combination is screened through a combination screening strategy, and each target field is taken as each hyperedge set; based on each hyperedge set, a sparse risk hypergraph is constructed; and weight values of each hyperedge set of the sparse risk hypergraph are identified. The desensitization module is configured to identify data types of each sensitive field in the enterprise data and desensitization influence categories of each sensitive field based on the sparse risk hypergraph and the weight values of each hyperedge set of the sparse risk hypergraph; based on the desensitization influence categories of each sensitive field and the data types of each sensitive field, a field desensitization cost algorithm is used to calculate field desensitization cost values corresponding to each sensitive field; based on the field desensitization cost values corresponding to each sensitive field, a desensitization coverage algorithm is used to identify each minimum desensitization coverage subset corresponding to the enterprise data in the enterprise data; and based on the data types of each sensitive field in each desensitization coverage subset, each minimum desensitization coverage subset is subjected to field desensitization processing through a desensitization strategy corresponding to each data type to obtain desensitized enterprise data. 5.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-4 when the computer program is executed by the processor. The processor executes the computer program to implement the steps of the method of any one of claims 1 to 3.

6. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 3.

7. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 3.

Images