Method, device and electronic equipment for determining risk level of sensitive data

Abstract

The application provides a method and device for determining the risk level of sensitive data and electronic equipment, the method comprising receiving a user data list; constructing a data object hierarchical model, wherein the data object hierarchical model is obtained using a plurality of sets of historical user data lists, each set of historical user data list in the plurality of sets of historical user data lists comprising the basic information and asset information of a plurality of users obtained in a historical time period; and using the data object hierarchical model to perform risk level assessment processing on the user data list to obtain the risk level of the user data list. The data ontology integrity dimension is introduced in the sensitive data grading method, the abstract degree of the sensitive data grading index is improved, and the comprehensiveness, accuracy of reflecting the data object and the corresponding entity object are improved, thereby solving the problem that the prior art cannot accurately reflect the true sensitivity of the data object and the corresponding entity object.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sensitive data detection processing, in particular to a sensitive data risk level determination method and device and electronic equipment. BACKGROUND

[0002] With the development of computer science, traditional boundary protection based on plugging has become unsustainable, and more and more information security products based on data security have emerged. The discovery of sensitive data has become a core technology related to product competitiveness. However, the current sensitive data classification is a simple and rough way based on keyword search and matching, which obviously does not meet the requirements of data security management for the purpose of accurate target protection, and it is inevitable to have false negatives and false positives.

[0003] Current technical solutions for sensitive data classification:

[0004] (1) Provide samples and define their data classification in an artificial experience judgment manner:

[0005] Various sample data are provided by data producers and users, and are classified and graded. In the case of using scenarios, especially using personnel, a data object can only belong to one security level.

[0006] (2) Learn sample data classification model in a machine learning manner:

[0007] In the natural language understanding layer, Chinese content recognition is more difficult than English content recognition, and the technology is more complex. Currently, domestic statistical analysis models based on Chinese semantics are mainly used for data discovery. Through Chinese word segmentation, a Chinese character sequence is segmented into individual words. By using statistical and machine learning methods, based on Chinese word properties and statistical characteristics, the probability of each segmented word appearing is calculated, and the probability of the most probable segmented word is found out quickly as the final result using the Viterbi algorithm.

[0008] (3) Define sensitive indicators:

[0009] Some keywords and attributes with the highest probability in the segmented results learned from sample data are used as quantitative indicators for sensitive degree classification, and rules for dividing levels are determined by combining statistical results. Key words, regular expression matching and other technical means are used to form sensitive data classification capabilities.

[0010] Disadvantages of the existing solution: The experience-based classification model often has experience bias or completeness vulnerabilities. Different business departments may have different object properties.

[0011] For example, data containing only customer name or address information can be considered as non-sensitive, but if the data object contains both name and address attributes, it should be considered as reflecting customer information and should be treated as sensitive data. If an account and amount attribute are added, it reflects transaction flow, and the sensitivity level should be increased.

[0012] The level division by counting only a single indicator is simple, but it is difficult to accurately reflect the true sensitivity of the data object and the corresponding entity object. SUMMARY

[0013] The main purpose of the present application is to provide a sensitive data risk level determination method and device and electronic equipment, so as to at least solve the problem that the prior art is difficult to accurately reflect the true sensitivity of the data object and the corresponding entity object.

[0014] In order to achieve the above purpose, according to one aspect of the present application, a sensitive data risk level determination method is provided, which comprises: receiving a user data list, the user data list comprising basic information and asset information of a plurality of users, the basic information comprising at least one of the following parameters: name, date of birth, gender, contact phone number, contact address, and the asset information comprising at least one of the following parameters: fixed asset information, non-fixed asset information; constructing a data object classification model, wherein the data object classification model is obtained using a plurality of sets of historical user data lists, each set of the historical user data lists comprising basic information and asset information of a plurality of users obtained in a historical time period; using the data object classification model to perform risk level assessment processing on the user data list to obtain the risk level of the user data list, the input of the data object classification model being the user data list, and the output of the data object classification model being the risk level of the user data list.

[0015] Optionally, using the data object classification model to perform risk level assessment processing on the user data list to obtain the risk level of the user data list comprises: using the data object classification model to obtain a sensitive zero integer ratio according to the basic information of the user data list; determining a sensitive weight according to the user data list; determining a sensitive score vector of the user data list according to the sensitive weight and the sensitive zero integer ratio; and determining the risk level of the user data list according to the sensitive score vector.

[0016] Optionally, the sensitive zero integer ratio is obtained according to the basic information of the user data list by using the data object hierarchical model, comprising: a feature matrix is obtained according to the basic information of the user data list by using the data object hierarchical model; a first total number and a second total number are determined according to the feature matrix, the first total number being a total number of parameters of the basic information of all the users in the user data list, and the second total number being a total number of each parameter of the basic information of all the users in the user data list; and the sensitive zero integer ratio is determined according to the first total number and the second total number.

[0017] Optionally, the feature matrix is obtained according to the basic information of the user data list by using the data object hierarchical model, comprising: in a case where the basic information of the user in the user data list comprises each parameter, corresponding parameters in the basic information of the user in the user data list are encoded according to a first numerical value; in a case where the basic information of the user in the user data list does not comprise each parameter, corresponding parameters in the basic information of the user in the user data list are encoded according to a second numerical value; and the feature matrix is generated according to all the first numerical values and all the second numerical values.

[0018] Optionally, the sensitive zero integer ratio is determined according to the first total number and the second total number, comprising: the sensitive zero integer ratio is determined according to the sensitive zero integer ratio is determined, wherein D i is the sensitive zero integer ratio of the i-th type information of all the users in the user data list, P j is a total number of the j-th parameter in the i-th type information of all the users in the user data list, P total is a total number of the i-th type information of all the users in the user data list, and m is a total number of types of information in the user data list.

[0019] Optionally, the sensitive score vector of the user data list is determined according to the sensitive weight and the sensitive zero integer ratio, comprising: the sensitive score vector of the user data list is determined according to the sensitive score vector of the user data list is determined, wherein W i is the sensitive weight of the i-th type information of all the users in the user data list, D i is the sensitive zero integer ratio of the i-th type information of all the users in the user data list, and Score is the sensitive score vector of the user data list, and n is a total number of all the users in the user data list.

[0020] Optionally, the risk level of the user data list is one of a plurality of risk levels, the sensitive score vector includes a plurality of probability values, each of the risk levels corresponds to one of the probability values, and determining the risk level of the user data list according to the sensitive score vector includes: determining a target probability value according to all of the probability values, the target probability value being a maximum value among all of the probability values; and determining, according to the target probability value, a risk level corresponding to the target probability value as the risk level of the user data list.

[0021] Optionally, the sensitive weight is determined according to the user data list by fitting the feature matrix using a convolutional neural network, the input of the convolutional neural network being the feature matrix, and the output of the convolutional neural network being the sensitive weight.

[0022] According to another aspect of the present application, a device for determining a risk level of sensitive data is provided, which includes a receiving unit, a constructing unit and a processing unit; the receiving unit is configured to receive a user data list, the user data list including basic information and asset information of a plurality of users, the basic information including at least one of the following parameters: name, date of birth, gender, contact number, contact address, and the asset information including at least one of the following parameters: fixed asset information, non-fixed asset information; the constructing unit is configured to construct a data object classification model, wherein the data object classification model is obtained using a plurality of sets of historical user data lists, each of the plurality of sets of historical user data lists including the basic information and asset information of a plurality of users obtained in a historical time period; and the processing unit is configured to perform risk level assessment processing on the user data list using the data object classification model to obtain a risk level of the user data list, the input of the data object classification model being the user data list, and the output of the data object classification model being the risk level of the user data list.

[0023] According to another aspect of the present application, an electronic device is provided, which includes one or more processors, memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include a program for performing any of the methods for determining a risk level of sensitive data.

[0024] By adopting the data object hierarchical model, the risk level of the user data list is obtained through risk level assessment processing of the user data list, the dimension of data ontology integrity is introduced in the sensitive data grading method, the abstract degree of the sensitive data grading index and the comprehensiveness and accuracy of the reflection of the entity object data attribute are improved, the model deviation and completeness vulnerability caused by the past simple counting of keywords and regular expressions and the experience-based grading method are made up, the classification and grading accuracy is improved, and the problem that the existing technology cannot accurately reflect the real sensitive degree of the data object and the corresponding entity object is solved. BRIEF DESCRIPTION OF DRAWINGS

[0025] The accompanying drawings constituting a part of the specification illustrate the present application and together with the description further serve to provide a further understanding of the present application. The schematic embodiments of the present application and the description thereof serve to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0026] Figure 1 A hardware structure block diagram of a mobile terminal for performing a sensitive data risk level determination method according to an embodiment of the present application is shown;

[0027] Figure 2 A flowchart of a sensitive data risk level determination method according to an embodiment of the present application is shown;

[0028] Figure 3 A structure block diagram of a sensitive data risk level determination apparatus according to an embodiment of the present application is shown.

[0029] Among the above drawings, the following reference signs are included:

[0030] 102, processor; 104, memory; 106, transmission device; 108, input and output device. DETAILED DESCRIPTION

[0031] It should be noted that the embodiments and features in the present application can be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0032] In order to enable persons skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by persons skilled in the art without creative labor should fall within the scope of protection of the present application.

[0033] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily have to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0034] For ease of description, the following describes some nouns or terms related to the embodiments of the present application:

[0035] Data classification: according to a specific classification rule, some documents in a specified storage on a local or network are classified, so that documents of the same level have the same or similar quantifiable characteristic values.

[0036] Ontology integrity: an abstract data object as a carrier container of a real-world entity object, the completeness of the attribute data of the real-world entity object contained in the data object.

[0037] Recurrent neural network (RNN): a type of recursive neural network that takes sequence data as input, recursively evolves in the direction of the sequence, and all nodes (recurrent units) are connected in a chain. Recurrent neural networks have applications in natural language processing (NLP), such as speech recognition, language modeling, machine translation, etc. They are also used for various time series forecasting.

[0038] Convolutional neural network (CNN): a type of feedforward neural network that contains convolutional calculations and has a deep structure, and is one of the representative algorithms of deep learning.

[0039] As described in the background section, existing experience-based grading models often suffer from experience bias or incompleteness. Different business departments may focus on different object attributes. For example, data containing only customer names or address information may be considered non-sensitive, but if a data object contains both name and address attributes, it should be considered to reflect customer information and should be treated as sensitive data. If account and amount attributes are added, it reflects transaction history, and the sensitivity level should be increased. Grading based solely on counting a single indicator is simplistic and fails to accurately reflect the true sensitivity of data objects and their corresponding entities. To address the problem that existing technologies cannot accurately reflect the true sensitivity of data objects and their corresponding entities, embodiments of this application provide a method, apparatus, and electronic device for determining the risk level of sensitive data.

[0040] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0041] The methods and embodiments provided in this application can be executed on a mobile terminal, computer terminal, or similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for a method of determining the risk level of sensitive data according to an embodiment of the present invention. Figure 1 As shown, a mobile terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the mobile terminal described above. For example, the mobile terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0042] The memory 104 can be used to store computer programs, such as software programs of application software and modules, such as a computer program corresponding to the device information display method of the embodiments of the present application. The processor 102 executes various functional applications and data processing, i.e., implements the above method, by running the computer programs stored in the memory 104. The memory 104 can include a high-speed random access memory, and can further include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 104 can further include memories remotely arranged with respect to the processor 102, which can be connected to the mobile terminal through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof. The transmission device 106 is used to receive or send data via a network. The specific examples of the above network can include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet in a wireless manner.

[0043] In the embodiments, a method for determining a risk level of sensitive data running on a mobile terminal, a computer terminal or a similar computing device is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0044] Figure 2 FIG. 1 is a flowchart of a method for determining a risk level of sensitive data according to an embodiment of the present application. As shown in FIG. 1, the method includes the following steps: Figure 2

[0045] In step S201, a user data list is received, the user data list including basic information and asset information of a plurality of users, the basic information including at least one of the following parameters: name, date of birth, gender, contact number, contact address, and the asset information including at least one of the following parameters: fixed asset information and non-fixed asset information.

[0046] The processing terminal retrieves user list data from a server, the fixed asset information including house information and vehicle information, and the non-fixed asset information including bank account information and bond information. The user data list is in a table format, and the basic information is not limited to the above information.​

[0047] In step S202, a data object hierarchical model is constructed, wherein the data object hierarchical model is obtained using a plurality of sets of historical user data lists, each of the plurality of sets of historical user data lists including basic information and asset information of a plurality of users obtained in a historical time period;

[0048] The historical user data list can be a historical user data list one month or one week ago, and part of the historical user data list is used as a training set and the other part is used as a test set to construct and test the data object hierarchical model. Documents are carriers of data, and data are carriers of information. Data objects are high-level semantics at the information level. The classification model based on high-level semantics has smaller granularity, is more accurate, and is more in line with the requirements of data classification and grading.

[0049] In step S203, the data object hierarchical model is used to perform risk level assessment processing on the user data list to obtain a risk level of the user data list. The input of the data object hierarchical model is the user data list, and the output of the data object hierarchical model is the risk level of the user data list.

[0050] In the above steps, the data object hierarchical model is used to perform risk level assessment processing on the user data list to obtain a risk level of the user data list. The data ontology integrity dimension is introduced into the sensitive data grading method, the abstraction degree of the sensitive data grading index and the comprehensiveness and accuracy of the reflection of the entity object data properties are improved, the model bias and completeness vulnerability caused by the past simple counting of keywords and regular expressions and the experience-based grading method are compensated, the classification and grading accuracy is improved, and the problem that the prior art cannot accurately reflect the true sensitivity of data objects and corresponding entity objects is solved.

[0051] Step S203 includes the following steps (i.e., using the data object hierarchical model to perform risk level assessment processing on the user data list to obtain a risk level of the user data list, including the following steps):

[0052] In step S2031, the data object hierarchical model is used to obtain a sensitive zero integer ratio according to the basic information of the user data list.

[0053] The sensitive zero integer ratio is used to reflect the sensitivity of data.

[0054] Step S2031 includes the following steps (i.e., using the data object hierarchical model to obtain a sensitive zero integer ratio according to the basic information of the user data list, including the following steps):

[0055] Step S20311, obtaining a feature matrix according to the basic information of the user data list by using the data object hierarchical model.

[0056] In an embodiment of the present application, obtaining a feature matrix according to the basic information of the user data list by using the data object hierarchical model comprises: in the case that the basic information of the user in the user data list includes parameters, encoding the corresponding parameters in the basic information of the user in the user data list according to a first numerical value; in the case that the basic information of the user in the user data list does not include parameters, encoding the corresponding parameters in the basic information of the user in the user data list according to a second numerical value; and generating the feature matrix according to all the first numerical values and all the second numerical values.

[0057] Specifically, as shown in Table 1, the first numerical value is 1 and the second numerical value is 0, i.e., in the case that the user A in the user data list includes a name, the name of the user A in the user data list is encoded according to 1.

[0058] Step S20312, determining a first total number and a second total number according to the feature matrix, wherein the first total number is the total number of parameters in the basic information of all the users in the user data list, and the second total number is the total number of parameters in the basic information of all the users in the user data list.

[0059] Step S20313, determining the sensitive zero integer ratio according to the first total number and the second total number.

[0060] Specifically, the feature matrix is shown in Table 1.

[0061] Table 1 Feature Matrix Table

[0062]

[0063] As shown in Table 1, the second total number is 2 in the case of referring to the total number of names, and the second total number is 2 in the case of referring to the total number of contact numbers, and other cases are not described here. In the case that there are only basic information and asset information of three users, the first total number is 6.

[0064] In an embodiment of the present application, determining the sensitive zero integer ratio according to the first total number and the second total number comprises:

[0065] According to determining the sensitive zero integer ratio, wherein D i is the sensitive zero integer ratio of the i-th type of information of all the users in the user data list, P jP is the total number of the jth parameter in the ith information of all the users in the user data list total m is the total number of the information categories in the user data list.

[0066] Specifically, the first information corresponding to the sensitive zero integer ratio is the basic information of the user, and the second information is the asset information of the user, for example, the total number of the jth parameter in the ith information of the user can be the total number of the name parameter of the basic information of the user, which is 2, the total number of the ith information of all the users in the user data list can be the total number of the basic information of the user, which is 3, and m is 5.

[0067] In step S2032, the sensitive weight is determined according to the user data list.

[0068] The sensitive weight is used to represent the proportion of each sensitive data;

[0069] In step S2033, the sensitive score vector of the user data list is determined according to the sensitive weight and the sensitive zero integer ratio.

[0070] The risk level of the user data list is one of a plurality of risk levels, and the sensitive score vector includes a plurality of probability values, each of the risk levels corresponding to one of the probability values.

[0071] In step S2034, the risk level of the user data list is determined according to the sensitive score vector.

[0072] Specifically, the data object classification model based on the integrity of the entity contained in the data object greatly improves the accuracy of sensitive data discovery and greatly reduces the information security events caused by false negatives and false positives. Based on the joint extraction model based on the recurrent neural network (the joint extraction model based on the recurrent neural network identifies the named entity and the entity relationship contained in the data object, obtains the entity and attribute list of the data object contained in the document, and the attribute list is shown in Table 1, including name, contact number, contact address, date of birth and gender), taking the data entity object contained in the document as the target and taking the data object integrity evaluation method as the core technology, the classification and grading of the document is realized.

[0073] In an embodiment of the present application, the risk level of the user data list is one of a plurality of risk levels, and the sensitive score vector includes a plurality of probability values, each of the risk levels corresponding to one of the probability values, and step S2034 includes the following steps (i.e., determining the risk level of the user data list according to the sensitive score vector, including the following steps):

[0074] Step S20341, according to all the above probability values, determine the target probability value, the target probability value is the maximum value in all the above probability values;

[0075] The target probability value facilitates subsequent determination of the risk level;

[0076] Step S20342, according to the target probability value, determine the risk level corresponding to the target probability value as the risk level of the user data list.

[0077] Specifically, the risk levels are respectively a first risk level, a second risk level, a third risk level and a fourth risk level, the risk levels increase in turn, the probability value of the first risk level is 18%, the probability value of the second risk level is 25%, the probability value of the third risk level is 35%, and the probability value of the fourth risk level is 22%, then 35% is the target probability value, and the probability of determining the third risk level is the risk level of the user data list.

[0078] The data object contained in the user data list is taken as a bearing container of the world entity object, a data object hierarchical model is established according to the completeness of the attribute data of the real world entity object contained in the data object, for example, a data object A containing only a name is not sensitive, but a data object B containing a name and a contact number is sensitive, for the data object A, the personnel entity only has a name, for the data object B, the personnel entity contains a name and a contact number, the completeness of the description of the personnel entity of the data object B is higher than that of the data object A, which is highly consistent with the requirement of the sensitive document classification.

[0079] Technically, the data object is converted into a mathematical matrix using an N-gram Chinese language model, and then the attributes and relationships of the entity object are extracted from the data object using a joint extraction model based on a recurrent neural network, the Chinese language model: the Chinese language model is a language abstract mathematical modeling according to the objective facts of Chinese language, the sentence is regarded as a word sequence (w1, w2, w3, …, wn), it is assumed that the Kth word is only related to the surrounding N words (generally, the value of N is 1, 3, 5, otherwise the calculation amount is too large), then when the first k (0<k<n) words appear, the probability P of the next word appearing is the N-gram Chinese language model.

[0080] According to the data standard, all types of data entities contained in the document are combined into a feature vector according to the sensitive zero integer ratio. As shown in Table 2:

[0081] Table 2: Feature vector composed of sensitive zero integer ratio

[0082] Basic information Asset information Contract information Account basic information Credit information 0.8 0.5 0.3 0.6 0.2

[0083] The sensitive zero ratio of the basic information is 0.8, the sensitive zero ratio of the asset information is 0.5, the sensitive zero ratio of the contract information is 0.3, the sensitive zero ratio of the account basic information is 0.6, the sensitive zero ratio of the credit information is 0.2, and the account basic information is the basic information of the bank account, such as the account number, the opening time, the deposit limit, the income and expenditure information and the like. The basic information is the basic information of the customer, for example, the name, the age, the gender, the contact information, the address and the like.

[0084] In an embodiment of the present application, according to the sensitive weight and the sensitive zero ratio, the sensitive score vector of the user data list is determined, including:

[0085] According to determining the sensitive score vector of the user data list, wherein W i is the sensitive weight of the i-th type information of all the users in the user data list, D i is the sensitive zero ratio of the i-th type information of all the users in the user data list, Score is the sensitive score vector of the user data list, and n is the total number of all the users in the user data list.

[0086] Specifically, the sensitive zero ratio of the i-th type information of all the users in the user data list and the sensitive weight of the i-th type information of all the users in the user data list are the same as the total number of the i-th type information of all the users in the user data list, which will not be repeated here.

[0087] In an embodiment of the present application, according to the user data list, the sensitive weight is determined, including: fitting the feature matrix by using a convolutional neural network to obtain the sensitive weight, the input of the convolutional neural network is the feature matrix, and the output of the convolutional neural network is the sensitive weight.

[0088] The corpus is selected from the mixed corpus, and the list of named entity categories to be labeled and the list of entity relationship categories are sorted according to the hierarchical data standard. Generally, the data classification is classified and graded according to the whole classification and grading of the documents containing sensitive content and non-sensitive content. The method of the present application is only for classifying and grading the sensitive content, and the granularity is different, which is more accurate.

[0089] The BRAT (an open-source text labeling system, named BRAT) corpus labeling system is built, and the corpus is named entity labeled using the BIO (a popular representation method of named entity labeling) label, mainly including the labeling of five categories of entities, i.e., person name, address name, organization name, time and number; the entity relationship is labeled using the BRAT system, mainly including six categories of entity relationship labels, i.e., social relationship of a person, physical position relationship, general membership relationship, whole and part relationship, organization membership relationship and possession relationship.

[0090] The embodiment of the present application further provides a device for determining the risk level of sensitive data. It should be noted that the device for determining the risk level of sensitive data can be used to execute the method for determining the risk level of sensitive data provided by the embodiment of the present application. The device is used to realize the above-mentioned embodiment and preferred embodiment, and will not be described here. As used below, the term "module" can be a combination of software and / or hardware that realizes a predetermined function. Although the device described in the following embodiment is preferably realized in software, the realization of hardware or a combination of software and hardware is also possible and is conceived.

[0091] The device for determining the risk level of sensitive data provided by the embodiment of the present application is introduced below.

[0092] Figure 3 is a structural block diagram of a device for determining the risk level of sensitive data provided by the embodiment of the present application. As shown in Figure 3 , the device includes a receiving unit 31, a constructing unit 32 and a processing unit 33; the receiving unit 31 is used to receive a user data list, the user data list including basic information and asset information of a plurality of users, the basic information including at least one of the following parameters: name, date of birth, gender, contact number, contact address, and the asset information including at least one of the following parameters: fixed asset information, non-fixed asset information; the constructing unit 32 is used to construct a data object hierarchical model, wherein the data object hierarchical model is obtained using a plurality of sets of historical user data lists, each of the plurality of sets of historical user data lists including the basic information and asset information of a plurality of users obtained in a historical time period; the processing unit 33 is used to perform risk level assessment processing on the user data list using the data object hierarchical model to obtain the risk level of the user data list, the input of the data object hierarchical model being the user data list, and the output of the data object hierarchical model being the risk level of the user data list.

[0093] In the device, the risk level of the user data list is obtained by using the data object hierarchical model to perform risk level assessment processing on the user data list, the dimension of data ontology integrity is introduced in the sensitive data grading method, the abstraction degree of the sensitive data grading index and the comprehensiveness and accuracy of the reflection of the entity object data attribute are improved, the model deviation and completeness vulnerability caused by the past simple counting of keywords and regular expressions and the experience-based grading method are compensated, the classification and grading accuracy is improved, and thus the problem that the existing technology cannot accurately reflect the real sensitivity of the data object and the corresponding entity object is solved.

[0094] In an embodiment of the present application, the processing unit includes a processing module, a first determination module, a second determination module, and a third determination module. The processing module is configured to obtain a sensitive zero-integral ratio according to the basic information of the user data list by using the data object hierarchical model. The first determination module is configured to determine a sensitive weight according to the user data list. The second determination module is configured to determine a sensitive score vector of the user data list according to the sensitive weight and the sensitive zero-integral ratio. The third determination module is configured to determine a risk level of the user data list according to the sensitive score vector.

[0095] In an embodiment of the present application, the processing module includes a first processing submodule, a second processing submodule, and a third processing submodule. The first processing submodule is configured to obtain a feature matrix according to the basic information of the user data list by using the data object hierarchical model. The second processing submodule is configured to determine a first total number and a second total number according to the feature matrix. The first total number is the total number of parameters of the basic information of all the users in the user data list. The second total number is the total number of each parameter of the basic information of all the users in the user data list. The third processing submodule is configured to determine the sensitive zero-integral ratio according to the first total number and the second total number.

[0096] In an embodiment of the present application, the first processing submodule includes a fourth processing submodule, a fifth processing submodule, and a sixth processing submodule. The fourth processing submodule is configured to encode the corresponding parameter in the basic information of the user in the user data list as a first numerical value in the case that the basic information of the user in the user data list includes each parameter. The fifth processing submodule is configured to encode the corresponding parameter in the basic information of the user in the user data list as a second numerical value in the case that the basic information of the user in the user data list does not include each parameter. The sixth processing submodule is configured to generate the feature matrix according to all the first numerical values and all the second numerical values.

[0097] In an embodiment of the present application, the third processing submodule comprises a first determining submodule;

[0098] The first determining submodule is configured to determine the sensitive weight of each type of information of each user in the user data list according to the feature matrix. Determine the sensitive zero integer ratio of the user data list, wherein W i D is the sensitive zero integer ratio of the i-th type of information of each user in the user data list, P j P is the total number of the j-th parameter in the i-th type of information of each user in the user data list, P total m is the total number of types of information in the user data list.

[0099] In an embodiment of the present application, the second determining module comprises a second determining submodule;

[0100] The second determining submodule is configured to determine the sensitive score vector of the user data list according to the feature matrix. Determine the sensitive score vector of the user data list, wherein W i D is the sensitive weight of the i-th type of information of each user in the user data list, D i D is the sensitive zero integer ratio of the i-th type of information of each user in the user data list, Score is the sensitive score vector of the user data list, and n is the total number of each user in the user data list.

[0101] In an embodiment of the present application, the risk level of the user data list is one of multiple risk levels, the sensitive score vector comprises multiple probability values, each risk level corresponds to a probability value, the third determining module comprises a third determining submodule and a fourth determining submodule, the third determining submodule is configured to determine a target probability value according to all the probability values, and the target probability value is the maximum value among all the probability values; and the fourth determining submodule is configured to determine the risk level corresponding to the target probability value as the risk level of the user data list according to the target probability value.

[0102] In an embodiment of the present application, the first determining module comprises a seventh processing submodule, the seventh processing submodule is configured to fit the feature matrix by using a convolutional neural network to obtain the sensitive weight, the input of the convolutional neural network is the feature matrix, and the output of the convolutional neural network is the sensitive weight.

[0103] The determination apparatus of the risk level of the sensitive data comprises a processor and a memory, the receiving unit, the constructing unit and the processing unit are all stored in the memory as program units, and the corresponding functions are realized by the processor executing the program units stored in the memory.

[0104] The processor comprises a core, and the core calls the corresponding program units in the memory.

[0105] The memory can comprise a non-permanent memory in a computer readable medium, a random access memory (RAM) and / or a non-volatile memory such as a read-only memory (ROM) or a flash memory (flash RAM), and the memory comprises at least one memory chip.

[0106] The embodiment of the present application provides a computer readable storage medium, the computer readable storage medium comprises a stored program, wherein the computer readable storage medium controls a device to execute the determination method of the risk level of the sensitive data when the program runs.

[0107] The embodiment of the present application provides a processor, the processor is used for running a program, wherein the processor executes the determination method of the risk level of the sensitive data when the program runs.

[0108] The embodiment of the present application provides a device, the device comprises a processor, a memory and a program stored in the memory and capable of running on the processor, and the processor executes the program to realize at least the following steps: receiving a user data list, the user data list comprising basic information and asset information of a plurality of users, the basic information comprising at least one of the following parameters: name, birth date, gender, contact number and contact address, and the asset information comprising at least one of the following parameters: fixed asset information and non-fixed asset information; constructing a data object hierarchical model, wherein the data object hierarchical model is obtained by using a plurality of groups of historical user data lists, each of the plurality of groups of historical user data lists comprises the basic information and the asset information of a plurality of users obtained in a historical time period; adopting the data object hierarchical model to perform risk level evaluation processing on the user data list to obtain the risk level of the user data list, the input of the data object hierarchical model being the user data list, and the output of the data object hierarchical model being the risk level of the user data list. The device in the present application can be a server, a PC, a PAD, a mobile phone and the like.

[0109] The application further provides a computer program product, which is suitable for executing a program including at least the following method steps when executed on a data processing device: receiving a user data list, wherein the user data list includes basic information and asset information of a plurality of users, the basic information includes at least one of the following parameters: name, date of birth, gender, contact phone number, contact address, and the asset information includes at least one of the following parameters: fixed asset information and non-fixed asset information; constructing a data object hierarchical model, wherein the data object hierarchical model is obtained by using a plurality of sets of historical user data lists, and each of the plurality of sets of historical user data lists includes the basic information and the asset information of a plurality of users obtained in a historical time period; and performing risk level assessment processing on the user data list by using the data object hierarchical model to obtain a risk level of the user data list, wherein the input of the data object hierarchical model is the user data list, and the output of the data object hierarchical model is the risk level of the user data list.

[0110] The application further provides an electronic device, which includes one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include a program for executing any one of the above-mentioned risk level determination methods of sensitive data.

[0111] Obviously, those skilled in the art should understand that the modules or steps of the application described above can be realized by general computing devices, which can be concentrated on a single computing device or distributed on a network composed of multiple computing devices, and can be realized by program codes executable by computing devices, so that they can be stored in storage devices and executed by computing devices, and in some cases, the steps shown or described can be executed in different orders, or they can be respectively manufactured into individual integrated circuit modules, or multiple modules or steps can be manufactured into a single integrated circuit module. Therefore, the application is not limited to any specific combination of hardware and software.

[0112] Those skilled in the art should understand that the embodiments of the application can be provided as methods, systems, or computer program products. Therefore, the application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the application can be in the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes.

[0113] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks.

[0114] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks.

[0115] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks.

[0116] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0117] The memory can include non-persistent memory, Random Access Memory (RAM), and / or non-volatile memory, e.g., Read Only Memory (ROM) or flash memory, among others. The memory is an example of computer-readable media.

[0118] Computer-readable media includes permanent and non-permanent, removable and non-removable media implemented in any method or technology for information storage. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carriers.

[0119] It should also be noted that the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, so that processes, methods, articles or devices including a series of elements not only include those elements, but also include other elements not explicitly listed or inherent to such processes, methods, articles or devices. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or device including the element.

[0120] From the above description, it can be seen that the above-mentioned embodiments of the present application achieve the following technical effects:

[0121] 1) The sensitive data risk level determination method of the present application, by using the above-mentioned data object classification model, carries out risk level evaluation processing on the user data list, obtains the risk level of the above-mentioned user data list, introduces the examination dimension of data ontology integrity in the sensitive data classification method, improves the abstraction degree of sensitive data classification index and the comprehensiveness and accuracy of reflecting the properties of entity object data, makes up for the model deviation and completeness vulnerability that may be caused by the past simple counting of keywords and regular expressions and the classification method based on experience, improves the classification accuracy, and solves the problem that the prior art cannot accurately reflect the true sensitivity of data objects and corresponding entity objects.

[0122] 2) The risk level determination device of sensitive data of the application, by using the above data object classification model, carries out risk level evaluation processing on the user data list, obtains the risk level of the above user data list, introduces the data ontology integrity investigation dimension in the sensitive data grading method, improves the abstract degree of the sensitive data grading index and the comprehensiveness, accuracy of reflecting the entity object data attribute, makes up the model deviation and completeness vulnerability caused by the past simple counting of keywords and regular expressions and the grading method based on experience, improves the classification and grading accuracy, and solves the problem that the prior art cannot accurately reflect the real sensitive degree of data objects and corresponding entity objects.

[0123] The above only describes the preferred embodiments of the application and is not intended to limit the application. Those skilled in the art can make various changes and modifications to the application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the application shall be included in the protection scope of the application.

Claims

1. A method for determining a risk level of sensitive data, characterized in that, The method comprises the following steps: receiving a user data list, the user data list comprising basic information and asset information of a plurality of users, the basic information comprising at least one of the following parameters: name, date of birth, gender, contact phone number, contact address, and the asset information comprising at least one of the following parameters: fixed asset information and non-fixed asset information; constructing a data object hierarchical model, wherein the data object hierarchical model is obtained using a plurality of sets of historical user data lists, each of the plurality of sets of historical user data lists comprising the basic information and asset information of a plurality of users obtained in a historical time period; performing risk level assessment processing on the user data list using the data object hierarchical model to obtain a risk level of the user data list, wherein the input of the data object hierarchical model is the user data list, and the output of the data object hierarchical model is the risk level of the user data list; obtaining a sensitive zero integer ratio from the basic information of the user data list using the data object hierarchical model, comprising: obtaining a feature matrix from the basic information of the user data list using the data object hierarchical model; determining a first total number and a second total number according to the feature matrix, wherein the first total number is the total number of parameters of the basic information of all users in the user data list, and the second total number is the total number of each parameter of the basic information of all users in the user data list; determining the sensitive zero integer ratio according to the first total number and the second total number; determining a sensitive weight from the user data list, comprising fitting the feature matrix using a convolutional neural network to obtain the sensitive weight, wherein the input of the convolutional neural network is the feature matrix, and the output of the convolutional neural network is the sensitive weight; performing risk level assessment processing on the user data list using the data object hierarchical model to obtain a risk level of the user data list, comprising: obtaining a sensitive zero integer ratio from the basic information of the user data list using the data object hierarchical model; determining a sensitive weight from the user data list; determining a sensitive score vector of the user data list according to the sensitive weight and the sensitive zero integer ratio; determining a risk level of the user data list according to the sensitive score vector.

2. The method of claim 1, wherein, obtaining a feature matrix from the basic information of the user data list using the data object hierarchical model, comprising: encoding corresponding parameters in the basic information of the user in the user data list according to a first numerical value in the case that the basic information of the user in the user data list comprises each parameter; encoding corresponding parameters in the basic information of the user in the user data list according to a second numerical value in the case that the basic information of the user in the user data list does not comprise each parameter; generating the feature matrix according to all the first numerical values and all the second numerical values.

3. The method of claim 1, wherein, According to the first total number and the second total number, the sensitive zero integer ratio is determined, comprising: according to Determine the sensitive zero-to-integrity ratio, wherein, For all the users in the user data list, the first The sensitive integer ratio of the class information, For all the users in the user data list, the first The first type of information The total number of parameters, For all the users in the user data list, the first The total number of information categories The total number of categories of information in the user data list.

4. The method of claim 1, wherein, According to the sensitive weight and the sensitive zero integer ratio, a sensitive score vector of the user data list is determined, comprising: according to Determine the sensitive rating vector of the user data list, where, For all the users in the user data list, the first The sensitivity weight of class information, For all the users in the user data list, the first The sensitive integer ratio of the class information, For the sensitive rating vector of the user data list, The total number of all users in the user data list.

5. The method of claim 1, wherein, The risk level of the user data list is one of a plurality of risk levels, the sensitive score vector includes a plurality of probability values, each of the risk levels corresponds to one of the probability values, according to the sensitive score vector, the risk level of the user data list is determined, comprising: According to all the probability values, a target probability value is determined, the target probability value is the maximum value of all the probability values; According to the target probability value, the risk level corresponding to the target probability value is determined as the risk level of the user data list.

6. A device for determining the risk level of sensitive data, characterized in that, Comprise: The receiving unit is used for receiving a user data list, the user data list includes basic information and asset information of a plurality of users, the basic information includes at least one of the following parameters: name, date of birth, gender, contact number, contact address, and the asset information includes at least one of the following parameters: fixed asset information, non-fixed asset information; The construction unit is used for constructing a data object hierarchical model, wherein the data object hierarchical model is obtained using a plurality of sets of historical user data lists, and each set of the historical user data lists includes basic information and asset information of a plurality of users obtained in a historical time period; The processing unit is used for performing risk level evaluation processing on the user data list by using the data object hierarchical model to obtain the risk level of the user data list, wherein the input of the data object hierarchical model is the user data list, and the output of the data object hierarchical model is the risk level of the user data list; The processing module includes a first processing submodule, a second processing submodule and a third processing submodule, the first processing submodule is used for obtaining a feature matrix according to the basic information of the user data list by using the data object hierarchical model; the second processing submodule is used for determining a first total number and a second total number according to the feature matrix, the first total number is the total number of parameters of the basic information of all users in the user data list, and the second total number is the total number of each parameter of the basic information of all users in the user data list; the third processing submodule is used for determining a sensitive zero integer ratio according to the first total number and the second total number; The first determining module includes a seventh processing submodule, the seventh processing submodule is used for fitting the feature matrix by using a convolutional neural network to obtain a sensitive weight, wherein the input of the convolutional neural network is the feature matrix, and the output of the convolutional neural network is the sensitive weight; The processing unit comprises a processing module, a first determining module, a second determining module and a third determining module. The processing module is configured to obtain a sensitive zero integer ratio according to basic information of the user data list by using a data object hierarchical model. The first determining module is configured to determine a sensitive weight according to the user data list. The second determining module is configured to determine a sensitive score vector of the user data list according to the sensitive weight and the sensitive zero integer ratio. The third determining module is configured to determine a risk level of the user data list according to the sensitive score vector.

7. An electronic device, comprising: The one or more processors, the memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs comprise a program for executing the method for determining the risk level of the sensitive data in any one of claims 1 to 5. ​

Images