Financial data desensitization method and device, electronic equipment, storage medium and product

Abstract

The invention discloses a financial data desensitization method and device, electronic equipment, a storage medium and a product, and relates to the technical field of financial science and technology. The method comprises the steps of determining a test data set in response to a desensitization instruction of a target data set, and classifying the test data set to obtain a financial test data set; determining a target desensitization rule matched with target financial data in the financial test data set, and desensitizing the target financial data based on the target desensitization rule; in response to a desensitization completion instruction of all target financial data in the financial test data set, determining whether the desensitization result of each target financial data meets a preset desensitization requirement; and under the condition of determining that each desensitization result meets a preset desensitization requirement, desensitizing each reference financial data in the target data set based on each target desensitization rule, and forwarding the desensitized target data set to a target analysis system. According to the scheme, the desensitization precision and efficiency of the amount data can be improved.

Description

Technical Field

[0001] The present invention relates to the field of financial technology, and in particular to a method, device, electronic device, storage medium and product for desensitizing financial data. Background Art

[0002] Financial data desensitization refers to the processing of financial data containing sensitive information to ensure that this information does not leak sensitive content such as personal identity information, account details, transaction records, etc. when used in a non-production environment.

[0003] At present, all acquired data is directly desensitized. When the data volume is large, it is difficult to quickly verify the desensitization results. At the same time, financial data and non-financial data are desensitized together, resulting in low desensitization accuracy of financial data.

[0004] How to improve the accuracy and efficiency of desensitizing monetary data is a major research issue in the industry. Summary of the invention

[0005] The present invention provides a method, device, electronic device, storage medium and product for desensitizing financial data to improve the desensitization accuracy and efficiency of monetary amount data.

[0006] According to one aspect of the present invention, a method for desensitizing financial data is provided, the method comprising:

[0007] In response to the desensitization instruction of the target data set, determine a test data set matching the target data set, and classify each data in the test data set to obtain a financial test data set and a non-financial test data set;

[0008] Determining a target desensitization rule that matches the target financial data in the financial test data set, and desensitizing the target financial data based on the target desensitization rule;

[0009] In response to a desensitization completion instruction for all target financial data in the financial test data set, determining whether a desensitization result of each target financial data meets a preset desensitization requirement;

[0010] When it is determined that the desensitization results of each of the target financial data meet the preset desensitization requirements, each reference financial data in the target data set is desensitized based on each of the target desensitization rules, and the desensitized target data set is forwarded to the target analysis system.

[0011] According to another aspect of the present invention, a financial data desensitization device is provided, the device comprising:

[0012] A test data set determination module, configured to determine a test data set matching the target data set in response to a desensitization instruction of the target data set, and classify each data in the test data set to obtain a financial test data set and a non-financial test data set;

[0013] a target desensitization rule determination module, configured to determine a target desensitization rule matching the target financial data in the financial test data set, and desensitize the target financial data based on the target desensitization rule;

[0014] a desensitization result verification module, configured to respond to a desensitization completion instruction for all target financial data in the financial test data set and determine whether a desensitization result of each target financial data satisfies a preset desensitization requirement;

[0015] The desensitization module is used to desensitize each reference financial data in the target data set based on each target desensitization rule when it is determined that the desensitization result of each target financial data meets the preset desensitization requirement, and forward the desensitized target data set to the target analysis system.

[0016] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0017] at least one processor; and

[0018] a memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for desensitizing financial data described in any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for desensitizing financial data described in any embodiment of the present invention when executed.

[0021] According to another aspect of the present invention, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the method for desensitizing financial data according to any embodiment of the present invention is implemented.

[0022] The technical solution of the embodiment of the present invention determines a test data set matching the target data set in response to a desensitization instruction of a target data set, and classifies each data in the test data set to obtain a financial test data set and a non-financial test data set; determines a target desensitization rule matching the target financial data in the financial test data set, and desensitizes the target financial data based on the target desensitization rule; responds to a desensitization completion instruction for all target financial data in the financial test data set, determines whether the desensitization result of each target financial data meets the preset desensitization requirement; when it is determined that the desensitization result of each target financial data meets the preset desensitization requirement, desensitizes each reference financial data in the target data set based on each target desensitization rule, and forwards the desensitized target data set to a target analysis system, which can improve the desensitization accuracy and efficiency of the amount data.

[0023] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 is a flow chart of a method for desensitizing financial data provided according to Embodiment 1 of the present invention;

[0026] Figure 2 is a flow chart of a method for desensitizing financial data provided according to Embodiment 2 of the present invention;

[0027] Figure 3 is a flow chart of a method for desensitizing financial data provided according to Embodiment 3 of the present invention;

[0028] Figure 4 is a flow chart of a method for desensitizing financial data provided according to a fourth embodiment of the present invention;

[0029] Figure 5 This is a schematic diagram of the structure of a financial data desensitization system provided according to a fourth embodiment of the present invention;

[0030] Figure 6 This is a schematic diagram of the structure of a financial data desensitization device provided according to Embodiment 5 of the present invention;

[0031] Figure 7 It is a schematic diagram of the structure of an electronic device for implementing the financial data desensitization method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0033] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0034] Embodiment 1

[0035] Figure 1 This is a flowchart of a method for desensitizing financial data according to the first embodiment of the present invention. This embodiment is applicable to desensitizing financial data. The method can be executed by a desensitizing device for financial data. The desensitizing device for financial data can be implemented in the form of hardware and / or software. The desensitizing device for financial data can be configured in electronic devices such as computers, servers or tablet computers. Figure 1 As shown, the method includes:

[0036] Step 110 : In response to the desensitization instruction of the target data set, determine a test data set matching the target data set, and classify each data in the test data set to obtain a financial test data set and a non-financial test data set.

[0037] The target data set may be consumption data, transfer data, remittance data, insurance purchase contract data or business data of each user, etc., which is not limited in this embodiment.

[0038] It is understandable that if the target data set contains sensitive data such as user information, specific transaction amounts in each transaction, financial statements or credit ratings, it is necessary to desensitize the target data set before sending it to the target analysis system to prevent leakage of sensitive information; wherein the target analysis system can be any data analysis system, for example, a target product recommendation system or a user determination system for the membership card level to be upgraded, etc., which is not limited in this embodiment.

[0039] It should be noted that, in this embodiment, the target data set is obtained after authorization by the user, and the acquisition method is reasonable and legal.

[0040] Optionally, in this embodiment, after receiving the desensitizing instruction for the target data set, a test data set that matches the target data set can be further determined; wherein, the test data set that matches the target data set can be the entire target data set, or it can be a partial target data set, that is, a part of the target data set; exemplarily, if the target data set contains 100,000 data items, then the test data set that matches the target data set can contain 2,000 data items; if the target data set contains 50,000 data items, then the test data set that matches the target data set can contain 500 data items; if the target data set contains 1,000 data items, then the test data set that matches the target data set can contain 1,000 data items.

[0041] In an optional implementation of this embodiment, after receiving the desensitizing instruction for the target data set, the amount of data contained in the target data set can be further determined. If the amount of data contained in the target data set is greater than a set quantity threshold (for example, 50,000, 100,000 or 1 million, etc.), then data in the target data set can be randomly selected according to the set comparison (for example, 1%, 2% or 10%, etc.) to obtain a test data set that matches the target data set.

[0042] Optionally, in this embodiment, after determining that a test data set matching the target data set is obtained, each data in the test data set may be further classified to obtain a financial test data set and a non-financial test data set.

[0043] In an optional implementation of this embodiment, after determining that a test data set is obtained that matches the target data set, the semantic information of each data in the test data set can be further determined. Furthermore, each data can be classified according to the semantic information of each data, thereby obtaining a financial test data set and a non-financial test data set.

[0044] In another optional implementation of the present embodiment, after determining that a test data set is obtained that matches the target data set, the test data set can be further input into a pre-trained machine learning model, for example, a classification model or a recognition model, which is not limited in the present embodiment; further, the classification result or recognition result of each data in the test data set can be obtained, thereby obtaining a financial test data set and a non-financial test data set.

[0045] Step 120: Determine a target desensitization rule that matches the target financial data in the financial test data set, and desensitize the target financial data based on the target desensitization rule.

[0046] Among them, the target financial data can be any data in the financial test data set, which is not limited in this embodiment; the target desensitization rules can be desensitization rules for financial data, such as numerical changes, character replacements, data obfuscation, pseudo-randomization or shielded display, etc., which are not limited in this embodiment.

[0047] In this embodiment, the target financial data may be any data in the financial test data set, which is not limited in this embodiment.

[0048] Optionally, in this embodiment, after classifying the test data set to obtain a financial test data set and a non-financial test data set, a target desensitization rule matching the target financial data can be further determined. Furthermore, the target financial data can be desensitized based on the target desensitization rule to obtain the desensitized target financial data.

[0049] In an optional implementation of the present embodiment, a target desensitization rule that matches the target financial data can be determined based on the data content of the target financial data; illustratively, if the data content of the target financial data is a specific sales amount, then the target desensitization rule can be a numerical transformation, for example, the sales amount can be subjected to a hash operation or a multiplication or division operation, etc.; if the data content of the target financial data is a bank account number, then the target desensitization rule can be a character replacement, for example, the bank account number can be replaced with “*”; if the data content of the target financial data is transaction data, then the target desensitization rule can be data obfuscation, for example, noise can be added to the target financial data.

[0050] Step 130: In response to the desensitization completion instruction for all target financial data in the financial test data set, determine whether the desensitization result of each target financial data meets the preset desensitization requirement.

[0051] Optionally, in this embodiment, after all the data in the financial test data set are desensitized based on the above steps and the desensitization results are obtained, each desensitization result can be further verified to determine whether each desensitization result meets the preset desensitization requirements; illustratively, it can be verified whether the desensitization result still contains sensitive information; if it is verified that the first financial data in the financial test data set contains sensitive information, for example, a bank card number, then it can be determined that the first financial data is not desensitized successfully, that is, each desensitization result of the financial test data set does not meet the preset desensitization requirements; if it is determined that all the data in the financial test data set does not contain sensitive information, then it can be determined that each desensitization result of the financial test data set meets the preset desensitization requirements.

[0052] Step 140: When it is determined that the desensitization result of each of the target financial data meets the preset desensitization requirement, desensitize each reference financial data in the target data set based on each of the target desensitization rules, and forward the desensitized target data set to the target analysis system.

[0053] The reference financial data in the target data set are all the financial data in the target data set. In this embodiment, for the convenience of description, they are named as reference financial data, which is not a limitation of this embodiment.

[0054] Optionally, in this embodiment, if it is determined that the masking results of each target financial data in the financial test data set meet the preset masking requirements, that is, each masking result does not contain sensitive information, then it can be determined that each target masking rule can meet the masking task of each data in the financial test data set, that is, it can meet the masking task of each financial data in the target data set; accordingly, in this embodiment, each reference financial data in the target data set can be further masked based on each target masking rule; exemplarily, the target masking rules matching each reference financial data can be determined respectively, and further, the first reference financial data can be masked based on the first target masking rule; wherein the first target masking rule is any masking rule; the first reference amount data is any reference financial data; the first target masking rule matches the first reference amount data.

[0055] Furthermore, the desensitized reference financial data can be forwarded to the target analysis system so that the target analysis system can analyze and process the reference financial data. In this way, the reference financial data received by the target analysis system does not contain sensitive information, thereby preventing the leakage of relevant information.

[0056] It should be noted that the solution of this embodiment first performs desensitization processing on the financial test data set with a smaller amount of data, so as to quickly obtain the desensitization results and quickly verify the accuracy of the desensitization results, and this process does not require human intervention; if the verification is passed, the financial data of the target data set continues to be desensitized, which can prevent the desensitization results from containing sensitive information, effectively improve the desensitization accuracy, and can improve the desensitization efficiency.

[0057] The solution of this embodiment determines a test data set matching the target data set in response to a desensitization instruction of the target data set, and classifies each data in the test data set to obtain a financial test data set and a non-financial test data set; determines a target desensitization rule matching the target financial data in the financial test data set, and desensitizes the target financial data based on the target desensitization rule; responds to a desensitization completion instruction for all target financial data in the financial test data set, determines whether the desensitization result of each target financial data meets the preset desensitization requirement; when it is determined that the desensitization result of each target financial data meets the preset desensitization requirement, desensitizes each reference financial data in the target data set based on each target desensitization rule, and forwards the desensitized target data set to the target analysis system, which can improve the desensitization accuracy and efficiency of financial data.

[0058] Embodiment 2

[0059] Figure 2 1 is a flowchart of a method for desensitizing financial data according to Embodiment 2 of the present invention. This embodiment is a further refinement of the above technical solution. The technical solution in this embodiment can be combined with each optional solution in one or more of the above embodiments. Figure 2 As shown, the method includes:

[0060] Step 210: In response to the desensitization instruction of the target data set, determine a test data set that matches the target data set, and classify each data in the test data set to obtain a financial test data set and a non-financial test data set.

[0061] Step 220: Determine a target desensitization rule that matches the target financial data in the financial test data set, and desensitize the target financial data based on the target desensitization rule.

[0062] Step 230: In response to the desensitization completion instruction for all target financial data in the financial test data set, determine whether the desensitization result of each target financial data meets the preset desensitization requirement.

[0063] Step 240: When it is determined that the first desensitization result of the first target financial data does not meet the preset desensitization requirement, generate a first desensitization rule matching the first target financial data based on a predetermined large model; desensitize the first target financial data based on the first desensitization rule to obtain a target first desensitization result; when it is determined that the first desensitization result meets the preset desensitization requirement, add the first desensitization rule to the target desensitization rule.

[0064] The first target financial data may be any financial data in the test financial test data set, which is not limited in this embodiment.

[0065] Optionally, in this embodiment, after desensitizing each financial data in the amount test data set, if it is determined that the first desensitization result of the first target financial data does not meet the preset desensitization requirement, that is, the first target financial data after desensitization still contains sensitive information, then a first desensitization rule matching the first target financial data can be generated based on the large model obtained by pre-fine-tuning; further, the first target financial data can be desensitized based on the first desensitization rule to obtain a first desensitization result of the first target financial data; further, it can be determined whether the first desensitization result meets the preset desensitization requirement, if the first desensitization result does not contain sensitive information, it can be determined that the first desensitization result meets the preset desensitization requirement; further, the first desensitization rule can be added to the target desensitization rule, so that subsequent financial data similar to the first target financial data can be desensitized by the first desensitization rule, which can effectively reduce the occurrence of desensitization results that do not meet the preset desensitization requirements.

[0066] In an optional implementation of the present embodiment, the large model can be obtained by fine-tuning multiple financial data containing sensitive information and the desensitization rules matching them. The specific fine-tuning process can be performed according to any fine-tuning method involved in the art. It is not described in detail in the present embodiment and is not a limitation of the present embodiment.

[0067] The solution of this embodiment can verify the desensitization result after desensitizing each financial data in the amount test data set, and after determining that the first desensitization result of the first target financial data does not meet the preset desensitization requirement, generate a first desensitization rule matching the first target financial data based on the large model obtained by pre-fine-tuning, and add the first desensitization rule to the target desensitization rule after the first desensitization result is verified, so as to update each target desensitization rule and provide a basis for improving the desensitization accuracy of financial data.

[0068] Embodiment 3

[0069] Figure 31 is a flowchart of a method for desensitizing financial data according to Embodiment 3 of the present invention. This embodiment is a further refinement of the above technical solution. The technical solution in this embodiment can be combined with each optional solution in one or more of the above embodiments. Figure 3 As shown, the method includes:

[0070] Step 310: In response to the desensitization instruction of the target data set, determine a test data set matching the target data set, and classify each data in the test data set to obtain a financial test data set and a non-financial test data set.

[0071] Optionally, in this embodiment, determining a test data set that matches the target data set, and classifying each data in the test data set to obtain a financial test data set and a non-financial test data set may include: determining the target data volume in the target data set; when it is determined that the target data volume is greater than a set data volume threshold, randomly extracting a set number of target data from the target data set to form the test data set; understanding each target data in the test data set to obtain content information of each target data; and classifying each target data based on the content information of each target data to obtain the financial test data set and the non-financial test data set.

[0072] Among them, the data volume threshold can be 50,000, 100,000 or 200,000, etc., which is not limited in this embodiment; the set number can be 500, 1,000 or 2,000, etc., which is not limited in this embodiment.

[0073] Optionally, in this embodiment, after receiving the desensitizing instruction regarding the target data set, the target data volume in the target data set can be further obtained, that is, the number of data contained in the target data set can be determined; further, the relationship between the target data volume and the data volume threshold can be determined. If it is determined that the target data volume is greater than the data volume threshold, for example, the target data volume is 50,000 and the data volume threshold is 10,000, then 500 target data can be randomly extracted from the target data set to form a test data set.

[0074] Furthermore, each target data in the test data set can be understood to obtain content information of each target data; further, each target data can be classified based on the content information of each target data to obtain a financial test data set and a non-financial test data set.

[0075] The content information of each target data may include: the type, specific meaning or value of the information carried by each target data, etc., which is not limited in this embodiment.

[0076] The solution of this embodiment extracts the test data set according to the amount of data in the target data set. When the target data set has a large amount of data, the desensitization result of a part of the data can be quickly obtained, which provides a basis for quickly verifying the desensitization result and ensures the desensitization accuracy of the financial data in the subsequent target data set.

[0077] Optionally, in this embodiment, after determining the target data volume in the target data set, the method may further include: if it is determined that the target data volume is less than or equal to a set data volume threshold, determining the target data set as a test data set.

[0078] In an optional implementation of this embodiment, if it is determined that the target data volume in the target data set is 600, the data volume threshold is 10,000, and the target data volume is much smaller than the data volume threshold, then there is no need to randomly extract the target data set, and the target data set can be directly determined as a test data set; further, it can be classified to obtain a financial test data set and a non-financial test data set.

[0079] The solution of this embodiment can directly classify the target data set when the target data set has a small amount of data, and perform subsequent desensitization processing based on the classification results, thereby improving the execution efficiency of the algorithm.

[0080] Step 320: input the target financial data into the pre-fine-tuned target macro model to obtain the target desensitization rule; process the target financial data according to the target desensitization rule to obtain the desensitization result of the target financial data.

[0081] Among them, the target big model is obtained by fine-tuning based on a pre-acquired first financial data set and a desensitizing target matching each financial data in the first financial data set; wherein, the first financial data set contains different categories of sensitive information related to finance, such as transaction amount, contract amount or user's account balance, etc., which is not limited in this embodiment.

[0082] Optionally, in this embodiment, after the test data set is classified into a financial test data set and a non-financial test data set, the target big model obtained by pre-fine-tuning each target financial data in the financial test data set can be used. The target big model determines a target desensitization rule that matches the target financial data based on the content information of the target financial data. Furthermore, the target financial data can be processed based on the target desensitization rule to obtain a desensitization result of the target financial data.

[0083] The advantage of this setting is that the desensitization results of various financial data can be obtained quickly and accurately, which helps to improve the desensitization accuracy of financial data.

[0084] It should be noted that, in this embodiment, the target desensitization rules can also be determined by the unfine-tuned large model. For example, the target desensitization rules can be obtained by setting appropriate prompt words; for example, the following content can be input into the unfine-tuned large model: The following is a column of data whose column name is transaction amount. Please help me choose the most appropriate desensitization rule from [Desensitization rule A, Desensitization rule B, Desensitization rule C,...].

[0085] Step 330: In response to the desensitization completion instruction for all target financial data in the financial test data set, determine whether the desensitization result of each target financial data meets the preset desensitization requirement.

[0086] Among them, the preset desensitization requirement may be a similarity requirement.

[0087] Correspondingly, in this embodiment, determining whether the desensitization results of each of the target financial data meet the preset desensitization requirements may include: performing similarity calculations on each of the desensitization results and each of the target financial data to obtain each similarity calculation result; if the target similarity calculation result is less than a set threshold, determining that the target desensitization result associated with the target similarity meets the desensitization requirements.

[0088] The threshold value may be set to a value such as 0.3, 0.4 or 0.5, which is not limited in this embodiment.

[0089] Optionally, in this embodiment, after obtaining the desensitized results of each target financial data in the amount test data set, the similarities between the desensitized results of each target financial data and each target financial data can be further calculated to obtain each similarity calculation result; if the target similarity calculation result is less than a set threshold, that is, the desensitized result of the target financial data and the target financial data have a low similarity, then it can be determined that the target financial data meets the desensitization requirements.

[0090] In another optional implementation of the present embodiment, if it is determined that the similarity between the desensitization result of the target financial data and the target financial data is 1, that is, the desensitization result of the target financial data is consistent with the target financial data, then it can be considered that the target financial data has not been desensitized successfully, and the desensitization result of the target financial data still contains sensitive information. At this time, it can be determined that the target financial data does not meet the desensitization requirements.

[0091] The solution of this embodiment can quickly determine whether the desensitization result of the target financial data meets the desensitization requirements through similarity calculation, thereby helping to improve the desensitization accuracy of the financial data.

[0092] Step 340: When it is determined that the desensitization results of each of the target financial data meet the preset desensitization requirements, desensitize each reference financial data in the target data set based on each of the target desensitization rules, and forward the desensitized target data set to the target analysis system.

[0093] The solution of this embodiment extracts the test data set according to the amount of data in the target data set. When the target data set has a large amount of data, the desensitization result of a part of the data can be quickly obtained, which provides a basis for quickly verifying the desensitization result and ensures the desensitization accuracy of the financial data in the subsequent target data set. At the same time, when the target data set has a small amount of data, the target data set can be directly classified, and subsequent desensitization processing can be performed based on the classification results, thereby improving the execution efficiency of the algorithm.

[0094] Embodiment 4

[0095] Figure 4 : is a flowchart of a method for desensitizing financial data provided according to Embodiment 4 of the present invention; this embodiment is a further refinement of the above technical solution, and the technical solution in this embodiment can be combined with each optional solution in one or more of the above embodiments. Figure 4 As shown, the method includes:

[0096] Step 410: In response to the desensitization instruction of the target data set, determine a test data set matching the target data set, and classify each data in the test data set to obtain a financial test data set and a non-financial test data set.

[0097] Step 420: determine the target data category of the target non-financial data according to the content information of the target non-financial data in the non-financial test data set; determine a reference desensitization rule matching the target data category in the target database, and desensitize the target non-financial data based on the reference desensitization rule.

[0098] The target non-financial data may be any non-financial data in the non-financial test data set, which is not limited in this embodiment.

[0099] The target data category may include at least one of the following: user name, user address, and user mobile phone number, etc., which are not limited in this embodiment; the reference desensitization rules that match the target data rules may be pseudo-randomization, display shielding, or character transformation, etc., which are not limited in this embodiment.

[0100] In this embodiment, the target database stores different data categories and reference desensitization rules corresponding to each data category. For example, the reference desensitization rule corresponding to the user name is pseudo-randomization, or the reference desensitization rule corresponding to the user address is character transformation.

[0101] Optionally, in this embodiment, after classifying the test data set to obtain a financial test data set and a non-financial test data set, the content information of each target non-financial data in the non-financial test data set can also be determined, and the target data category of the target non-financial data can be determined based on the content information. Furthermore, a reference desensitization rule corresponding to the target data category can be determined in the target database, and further, the target non-financial data can be desensitized based on the reference desensitization rule.

[0102] The solution of this embodiment, after classifying the test data set to obtain the financial test data set and the non-financial test data set, can also determine the content information of each target non-financial data in the non-financial test data set, and determine the target data category of the target non-financial data based on the content information. Further, a reference desensitization rule corresponding to the target data category can be determined in the target database. Further, the target non-financial data can be desensitized based on the reference desensitization rule, and the non-financial data can be desensitized quickly and accurately, which helps to obtain non-sensitive data.

[0103] In order to better understand the desensitization method of financial data involved in this embodiment, Figure 5 is a schematic diagram of the structure of a financial data desensitization system provided according to Embodiment 4 of the present invention. Figure 5 As shown, the financial data desensitization system includes: a core control module 510, a large model-natural language processing module 520, a large model-natural language processing module 530, a large model-financial data desensitization module 540 and a data processing module 550.

[0104] Core control module 510: can be used to provide page services, receive user input, control and coordinate the overall operation of the system, and manage the overall system functions, large model training and reasoning processes.

[0105] The big model-natural language processing module 520 processes the natural language requirements of the user during use based on the big model technology, converts them into commands that can be recognized by the system and provides them to the core control module for execution.

[0106] The large model-natural language processing module 530 can identify sensitive information in the data based on the input data to be desensitized. For financial numerical data, it is sent to the large model-financial data desensitization module 540 for independent processing; for other types of sensitive data, such as residential address, ID number, name and telephone number, the model determines and selects the existing desensitization rules.

[0107] The big model-financial data desensitization module 540 can generate desensitization rules for financial numerical data based on the input user needs and data types. For example, for the total sales and sales volume of a certain product, it is necessary to convert the specific values ​​into estimated values ​​while ensuring the accuracy of the qualitative analysis results, so as to confuse the pricing of the specific product.

[0108] The data processing module 550 can be used to accept desensitization rules to desensitize the data, store the original data, desensitized data and model parameters.

[0109] For example, in this embodiment, the process of the financial data desensitization system in the use phase may be as follows:

[0110] Step 1: Data import and preprocessing: The original data is imported from the external source into the big model-financial data desensitization module 540 for storage and undergoes preprocessing steps, including data cleaning, format standardization, etc., to ensure the quality and consistency of the data.

[0111] Step 2: The user uses natural language to propose a desensitization requirement for a certain data set; specifically, the core control module 510 uses the large model-natural language processing module 520 to understand the natural language and obtain corresponding operation commands; further, the data is sent to the large model-natural language processing module 530 for processing.

[0112] Step 3: The large model-natural language processing module 530 uses the large model to judge the type of input data and identify financial numerical and other types of data columns. For columns of non-financial numerical types, the corresponding data desensitization rules are matched in the stock desensitization rules for desensitization, and the results are sent to the data processing module 550; for columns of financial numerical types, they are sent to the large model-financial data desensitization module 540 for processing.

[0113] Step 4: The big model-financial data desensitization module 540 generates desensitization rules for the columns of input financial numerical types using the big model according to the desensitization requirements proposed by the user, and sends the results to the data processing module 550.

[0114] Step 5: After receiving the desensitization rules sent by the large model-natural language processing module 530 and the large model-financial data desensitization module 540, a small amount of data is desensitized for user preview. After seeing the desensitization results, the user can manually switch the selection of the existing desensitization rules or issue a new desensitization requirement to change the desensitization rules (for example, customized desensitization such as difference preservation and synthetic data is required).

[0115] Step 6: When there is a need to adjust the desensitization rules, repeat steps 3-5 until the user finally confirms that no adjustment is required. If no adjustment is required, the data processing module 550 desensitizes and stores the complete data according to the specified desensitization rules; the user can download the desensitized data for data analysis or other business purposes, such as conducting targeted customer marketing, external product communication and display, and regional product activities.

[0116] The solution of this embodiment realizes the isolation between users and data. Users always see the desensitized data, which ensures the security of data.

[0117] Embodiment 5

[0118] Figure 6 Schematic diagram of a financial data desensitization device according to Embodiment 5 of the present invention. Figure 6 As shown, the device includes: a test data set determination module 610, a target desensitization rule determination module 620, a desensitization result verification module 630 and a desensitization module 640.

[0119] The test data set determination module 610 is used to determine a test data set matching the target data set in response to the desensitization instruction of the target data set, and classify each data in the test data set to obtain a financial test data set and a non-financial test data set;

[0120] A target desensitization rule determination module 620, configured to determine a target desensitization rule that matches the target financial data in the financial test data set, and desensitize the target financial data based on the target desensitization rule;

[0121] A desensitization result verification module 630, configured to respond to a desensitization completion instruction for all target financial data in the financial test data set and determine whether a desensitization result of each target financial data satisfies a preset desensitization requirement;

[0122] The desensitizing module 640 is used to desensitize each reference financial data in the target data set based on each target desensitizing rule when it is determined that the desensitization result of each target financial data meets the preset desensitization requirement, and forward the desensitized target data set to the target analysis system.

[0123] In the scheme of this embodiment, the test data set determination module responds to the desensitization instruction of the target data set to determine the test data set matching the target data set, and classifies each data in the test data set to obtain a financial test data set and a non-financial test data set; the target desensitization rule determination module determines the target desensitization rule matching the target financial data in the financial test data set, and desensitizes the target financial data based on the target desensitization rule; the desensitization result verification module responds to the desensitization completion instruction of all the target financial data in the financial test data set to determine whether the desensitization result of each target financial data meets the preset desensitization requirement; the desensitization module, when determining that the desensitization result of each target financial data meets the preset desensitization requirement, desensitizes each reference financial data in the target data set based on each target desensitization rule, and forwards the desensitized target data set to the target analysis system, which can improve the desensitization accuracy and efficiency of the amount data.

[0124] In an optional implementation of this embodiment, the financial data desensitization device further includes: a desensitization rule generation module, which is used to generate a first desensitization rule matching the first target financial data based on a predetermined large model when it is determined that the first desensitization result of the first target financial data does not meet the preset desensitization requirement;

[0125] Desensitize the first target financial data based on the first desensitization rule to obtain a target first desensitization result;

[0126] When it is determined that the first desensitization result meets the preset desensitization requirement, the first desensitization rule is added to the target desensitization rule.

[0127] In an optional implementation of this embodiment, the test data set determination module 610 is specifically used to determine the target data volume in the target data set;

[0128] When it is determined that the target data volume is greater than a set data volume threshold, randomly extracting a set number of target data from the target data set to form the test data set;

[0129] Understanding each target data in the test data set to obtain content information of each target data;

[0130] The target data are classified based on the content information of each target data to obtain the financial test data set and the non-financial test data set.

[0131] In an optional implementation of this embodiment, the test data set determination module 610 is further specifically configured to determine the target data set as the test data set when it is determined that the target data amount is less than or equal to a set data amount threshold.

[0132] In an optional implementation of this embodiment, the financial data desensitization device further includes: a non-financial data desensitization module, configured to determine a target data category of the target non-financial data according to content information of the target non-financial data in the non-financial test data set;

[0133] A reference desensitization rule matching the target data category is determined in a target database, and the target non-financial data is desensitized based on the reference desensitization rule.

[0134] In an optional implementation of this embodiment, the target desensitization rule determination module 620 is specifically used to input the target financial data into the pre-fine-tuned target macro model to obtain the target desensitization rule;

[0135] Processing the target financial data according to the target desensitization rule to obtain a desensitization result of the target financial data;

[0136] The target large model is obtained by fine-tuning based on a pre-acquired first financial data set and a desensitization target matching each financial data in the first financial data set.

[0137] In an optional implementation of this embodiment, the preset desensitization requirement is a similarity requirement;

[0138] Correspondingly, the desensitization result verification module 630 is specifically used to perform similarity calculation between each desensitization result and each target financial data to obtain each similarity calculation result;

[0139] If the target similarity calculation result is less than the set threshold, it is determined that the target desensitization result associated with the target similarity meets the desensitization requirement.

[0140] The financial data desensitization device provided in the embodiment of the present invention can execute the financial data desensitization method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0141] In the technical solutions of the embodiments of the present invention, the collection, storage, use, processing, transmission, provision and disclosure of user personal information (such as name, residential address or consumption information, etc.) are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0142] Embodiment 6

[0143] Figure 7A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0144] like Figure 7 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0145] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0146] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a desensitization method for financial data, which includes: responding to a desensitization instruction of a target data set, determining a test data set that matches the target data set, and classifying each data in the test data set to obtain a financial test data set and a non-financial test data set; determining a target desensitization rule that matches the target financial data in the financial test data set, and desensitizing the target financial data based on the target desensitization rule; responding to a desensitization completion instruction for all target financial data in the financial test data set, determining whether the desensitization result of each target financial data meets a preset desensitization requirement; when it is determined that the desensitization result of each target financial data meets the preset desensitization requirement, desensitizing each reference financial data in the target data set based on each target desensitization rule, and forwarding the desensitized target data set to a target analysis system.

[0147] In some embodiments, the method for desensitizing financial data may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the method for desensitizing financial data described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the method for desensitizing financial data in any other appropriate manner (e.g., by means of firmware).

[0148] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0149] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0150] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0151] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0152] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0153] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0154] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0155] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

[0156] Embodiment 7

[0157] An embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the database detection method provided in any embodiment of the present application.

[0158] In the process of implementation, the computer program product can be written in one or more programming languages ​​or a combination thereof to perform the computer program code of the present invention, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0159] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0160] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for desensitizing financial data, characterized in that: include: In response to the desensitization instruction of the target data set, determine a test data set matching the target data set, and classify each data in the test data set to obtain a financial test data set and a non-financial test data set; Determining a target desensitization rule that matches the target financial data in the financial test data set, and desensitizing the target financial data based on the target desensitization rule; In response to a desensitization completion instruction for all target financial data in the financial test data set, determining whether a desensitization result of each target financial data meets a preset desensitization requirement; When it is determined that the desensitization results of each of the target financial data meet the preset desensitization requirements, each reference financial data in the target data set is desensitized based on each of the target desensitization rules, and the desensitized target data set is forwarded to the target analysis system.

2. The method for desensitizing financial data according to claim 1, characterized in that: The method further comprises: In the case where it is determined that the first desensitization result of the first target financial data does not meet the preset desensitization requirement, generating a first desensitization rule matching the first target financial data based on a predetermined macro model; Desensitize the first target financial data based on the first desensitization rule to obtain a target first desensitization result; When it is determined that the first desensitization result meets the preset desensitization requirement, the first desensitization rule is added to the target desensitization rule.

3. The method for desensitizing financial data according to claim 1, characterized in that: The determining of a test data set matching the target data set and classifying each data in the test data set to obtain a financial test data set and a non-financial test data set includes: Determining a target data volume in the target data set; When it is determined that the target data volume is greater than a set data volume threshold, randomly extracting a set number of target data from the target data set to form the test data set; Understanding each target data in the test data set to obtain content information of each target data; The target data are classified based on the content information of each target data to obtain the financial test data set and the non-financial test data set.

4. The method for desensitizing financial data according to claim 3, characterized in that: After determining the target data volume in the target data set, the method further includes: When it is determined that the target data volume is less than or equal to the set data volume threshold, the target data set is determined as a test data set.

5. The method for desensitizing financial data according to claim 3, characterized in that: After obtaining the financial test data set and the non-financial test data set, it also includes: Determining a target data category of the target non-financial data according to content information of the target non-financial data in the non-financial test data set; A reference desensitization rule matching the target data category is determined in a target database, and the target non-financial data is desensitized based on the reference desensitization rule.

6. The method for desensitizing financial data according to claim 1, characterized in that: The determining a target desensitization rule matching the target financial data in the financial test data set, and desensitizing the target financial data based on the target desensitization rule, includes: Inputting the target financial data into the pre-fine-tuned target macro model to obtain the target desensitization rules; Processing the target financial data according to the target desensitization rule to obtain a desensitization result of the target financial data; The target large model is obtained by fine-tuning based on a pre-acquired first financial data set and a desensitization target matching each financial data in the first financial data set.

7. The method for desensitizing financial data according to claim 1, characterized in that: The preset desensitization requirement is a similarity requirement; Accordingly, the step of determining whether the desensitization result of each target financial data meets the preset desensitization requirement includes: Performing similarity calculations on each of the desensitized results and each of the target financial data to obtain each similarity calculation result; If the target similarity calculation result is less than the set threshold, it is determined that the target desensitization result associated with the target similarity meets the desensitization requirement.

8. A desensitizing device for financial data, characterized in that: include: A test data set determination module, configured to determine a test data set matching the target data set in response to a desensitization instruction of the target data set, and classify each data in the test data set to obtain a financial test data set and a non-financial test data set; a target desensitization rule determination module, configured to determine a target desensitization rule matching the target financial data in the financial test data set, and desensitize the target financial data based on the target desensitization rule; a desensitization result verification module, configured to respond to a desensitization completion instruction for all target financial data in the financial test data set and determine whether a desensitization result of each target financial data satisfies a preset desensitization requirement; The desensitization module is used to desensitize each reference financial data in the target data set based on each target desensitization rule when it is determined that the desensitization result of each target financial data meets the preset desensitization requirement, and forward the desensitized target data set to the target analysis system.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the method for desensitizing financial data according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for desensitizing financial data according to any one of claims 1 to 7 when executed.

11. A computer program product, comprising a computer program, which, when executed by a processor, implements the method for desensitizing financial data according to any one of claims 1 to 7.

Images