Risk assessment method and device, and computer storable medium

Abstract

The present disclosure relates to a risk assessment method and device, a computer storable medium, and to the field of network security technology. The risk assessment method includes: obtaining multiple target uniform resource locators (URLs) corresponding to a target system; parsing one or more specified text contents in a target web page corresponding to each target URL to obtain one or more target text vectors, one specified text content corresponding to one target text vector; for each target URL, determining the privacy type of each target text vector; for each target URL, determining the risk value of the one or more specified text contents causing page privacy leakage in the target system according to the privacy type of one or more target text vectors. According to the present disclosure, the efficiency and accuracy of risk assessment can be improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of network security, and in particular to a risk assessment method and device, and a computer storable medium. Background Art

[0002] In order to meet the diverse needs of business, the operator system has developed and used various business operation systems, which contain a large number of front-end Web (World Wide Web) pages and display a large amount of personal information to users. The personal information displayed on the front-end Web pages of these systems may not be controlled in accordance with compliance requirements.

[0003] In the related art, whether page privacy leakage may occur in the target system is determined by manual inspection or semi-manual inspection. Summary of the invention

[0004] The inventor believes that the related technology uses manual or semi-manual methods to perform risk assessment, and the efficiency and accuracy of risk assessment are low.

[0005] In response to the above technical problems, the present disclosure proposes a solution that can improve the efficiency and accuracy of risk assessment.

[0006] According to a first aspect of the present disclosure, a risk assessment method is provided, comprising: obtaining a plurality of target uniform resource locators (URLs) corresponding to a target system; parsing one or more specified text contents in a target web page corresponding to each target URL to obtain one or more target text vectors, one specified text content corresponding to one target text vector; for each target URL, determining a privacy type of each target text vector; for each target URL, determining, based on the privacy type of the one or more target text vectors, a risk value of page privacy leakage in the target system caused by the one or more specified text contents.

[0007] In some embodiments, the one or more specified text contents are multiple, the one or more target text vectors are multiple, the multiple target text vectors include at least one first target text vector with a privacy type of a first privacy type, the first privacy type includes multiple first subtypes, and determining the risk value of the one or more specified text contents causing page privacy leakage in the target system includes: determining the similarity between the first target text vector space and the reference privacy feature vector space based on the first subtype of the at least one first target text vector, the first target text vector space including the at least one first target text vector, and the reference privacy feature vector space including one or more reference privacy feature vectors; determining the risk value of the specified text content corresponding to the at least one first target text vector causing page privacy leakage in the target system based on the similarity, the risk value being positively correlated with the similarity.

[0008] In some embodiments, determining the similarity between the first target text vector space and the reference privacy feature vector space based on the first subtype of the at least one first target text vector includes: obtaining a type weight value corresponding to the first subtype of the at least one first target text vector; determining the similarity between the first target text vector space and the reference privacy feature vector space based on the frequency of occurrence of the at least one first target text vector in the target web page and at least one type weight value.

[0009] In some embodiments, determining the similarity between the first target text vector space and the reference privacy feature vector space according to the frequency of occurrence of the at least one first target text vector in the target web page and at least one type weight value includes:

[0010] For each first target text vector, determine the sensitivity weight value of each first target text vector based on the frequency of occurrence of the at least one first target text vector in the target web page and the at least one type weight value, wherein the sensitivity weight value is positively correlated with the privacy sensitivity of each first target text vector; determine the similarity between the first target text vector space and the reference privacy feature vector space based on the sensitivity weight value of the at least one first target text vector.

[0011] In some embodiments, there are multiple at least one first target text vector, and determining the sensitivity weight value of each first target text vector includes: for each first target text vector, determining the sensitivity weight value of each first target text vector based on the sum of the squares of the products of the frequencies of the respective first target text vectors and the type weight values, and the product of the frequencies of each first target text vector and the type weight values.

[0012] In some embodiments, there are multiple at least one first target text vector and multiple one or more reference privacy feature vectors. Determining the similarity between the first target text vector space and the reference privacy feature vector space based on the sensitivity weight value of the at least one first target text vector includes: calculating the weighted sum of the multiple first target text vectors using the sensitivity weight values ​​of the multiple first target text vectors; and determining the similarity between the first target text vector space and the reference privacy feature vector space based on the weighted sum and the sum of the multiple reference privacy feature vectors.

[0013] In some embodiments, determining the similarity between the first target text vector space and the reference privacy feature vector space based on the weighted sum and the sum of multiple reference privacy feature vectors includes: determining the similarity between the first target text vector space and the reference privacy feature vector space based on the product of the weighted sum and the sum of multiple reference privacy feature vectors, the sum of the squares of the multiple reference privacy feature vectors, and the sum of the squares of the products of each first target text vector and the corresponding sensitivity weight value.

[0014] In some embodiments, determining, based on the similarity, the risk value of page privacy leakage occurring in the target system due to the specified text content corresponding to each of the first target text vectors includes: determining, based on the similarity and the current system permission value, the risk value of page privacy leakage occurring in the target system due to the specified text content corresponding to each of the first target text vectors, wherein the current system permission value is positively correlated with the size of permissions for accessing the target system.

[0015] In some embodiments, determining, based on the similarity and the system permission value, the risk value of page privacy leakage in the target system caused by the specified text content corresponding to the at least one first target text vector includes: determining a target permission value based on the current system permission value and the system permission discrete rate, the target permission value being positively correlated with the current system permission value and the system permission discrete rate, respectively, and the system permission discrete rate being used to expand the distinction between different system permission values; determining, based on the similarity and the target permission value, the risk value of page privacy leakage in the target system caused by the specified text content corresponding to the at least one first target text vector, the risk value being negatively correlated with the target permission value.

[0016] In some embodiments, the risk assessment method also includes: crawling multiple reference text contents of a specified web page; using the Gensim model to obtain the first subtype of each reference text content and the correlation between the reference text content and the first subtype; and determining the text vectors of one or more reference text contents whose correlation is greater than a correlation threshold as the reference privacy feature vector.

[0017] In some embodiments, the one or more target text vectors are multiple, and the multiple target text vectors include at least one second target text vector whose privacy type is a second privacy type. Determining the risk value of the one or more specified text contents causing the target system to have page privacy leakage includes: determining the target permission value according to the current system permission value and the system permission discrete rate, the current system permission value is positively correlated with the size of the permission to access the target system, the target permission value is positively correlated with the current system permission value and the system permission discrete rate, respectively, and the system permission discrete rate is used to expand the distinction between different system permission values; determining the risk value of the target system to have page privacy leakage caused by the specified text content corresponding to the at least one second target text vector according to the target permission value, and the risk value is negatively correlated with the target permission value.

[0018] In some embodiments, obtaining multiple target uniform resource locators URLs corresponding to the target system includes: using preset identity information to crawl one or more web pages related to one or more preset seed URLs; determining a target URL queue based on the relevance of the one or more web pages to the one or more preset seed URLs, the target URL queue including multiple target URLs arranged in descending order according to the relevance, and the multiple target URLs are at least a part of the URLs of the one or more web pages.

[0019] In some embodiments, parsing one or more specified text contents in the target web page corresponding to each target URL to obtain one or more target text vectors includes: according to the first-in-first-out rule of the target URL queue, parsing one or more specified text contents in the target web page corresponding to each target URL in turn to obtain one or more target text vectors.

[0020] In some embodiments, using preset identity information to crawl one or more web pages related to one or more preset seed URLs includes: using preset identity information to crawl one or more web pages related to one or more preset seed URLs that meet preset crawler conditions until the crawling time is equal to the preset crawler time.

[0021] According to a second aspect of the present disclosure, a risk assessment device is provided, comprising: an acquisition module, configured to acquire multiple target uniform resource locators (URLs) corresponding to a target system; a parsing module, configured to parse one or more specified text contents in a target web page corresponding to each target URL to obtain one or more target text vectors, one specified text content corresponding to one target text vector; a first determination module, configured to determine, for each target URL, a privacy type of each target text vector; and a second determination module, configured to determine, for each target URL, a risk value of page privacy leakage of the target system caused by the one or more specified text contents according to the privacy type of the one or more target text vectors.

[0022] According to a third aspect of the present disclosure, a risk assessment device is provided, comprising: a memory; and a processor coupled to the memory, wherein the processor is configured to execute the risk assessment method described in any of the above embodiments based on instructions stored in the memory.

[0023] According to a fourth aspect of the present disclosure, a computer storable medium is provided, on which computer program instructions are stored, and when the instructions are executed by a processor, the risk assessment method described in any of the above embodiments is implemented.

[0024] In the above embodiments, the efficiency and accuracy of risk assessment can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings, which constitute a part of the specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0026] The present disclosure may be more clearly understood from the following detailed description with reference to the accompanying drawings, in which:

[0027] Figure 1 is a flow chart illustrating a risk assessment method according to some embodiments of the present disclosure;

[0028] Figure 2 is a schematic diagram showing a tag tree of a certain target URL according to some embodiments of the present disclosure;

[0029] Figure 3 is a flow chart showing a method of determining a risk value of page privacy leakage in a target system caused by one or more specified text contents according to some embodiments of the present disclosure;

[0030] Figure 4 is a flowchart illustrating determining a similarity between a first target text vector space and a reference privacy feature vector space according to a first subtype of at least one first target text vector according to some embodiments of the present disclosure;

[0031] Figure 5 is a block diagram illustrating a risk assessment system according to some embodiments of the present disclosure;

[0032] Figure 6 is a block diagram illustrating a risk assessment apparatus according to some embodiments of the present disclosure;

[0033] Figure 7 is a block diagram showing a risk assessment device according to some other embodiments of the present disclosure

[0034] Figure 8 is a block diagram illustrating a computer system for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0035] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure unless otherwise specifically stated.

[0036] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0037] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0038] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0039] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0040] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0041] Figure 1 is a flow chart illustrating a risk assessment method according to some embodiments of the present disclosure.

[0042] like Figure 1 As shown, the risk assessment method includes steps S10 to S70. In some embodiments, the risk assessment method is performed by a risk assessment device.

[0043] In step S10, a plurality of target URLs (Uniform Resource Locators) corresponding to the target system are obtained.

[0044] In some embodiments, the preset identity information can be used to crawl one or more web pages related to one or more preset seed URLs, and the target URL queue is determined according to the relevance of the one or more web pages to the one or more preset seed URLs. The target URL queue includes multiple target URLs arranged in descending order according to the relevance. The multiple target URLs are at least a part of the URLs of one or more web pages. This process is a conventional step of the improved Fish-Search algorithm and will not be repeated here.

[0045] In some embodiments, preset identity information can be used to crawl one or more web pages related to one or more preset seed URLs and satisfying preset crawler conditions until the crawling duration is equal to the preset crawler duration. For example, the preset identity information includes a user name and password for logging into the target system. The preset crawler conditions include that the crawled web pages include specified keywords or that the domain name of the crawled web pages is a domain name below a specified second-level domain name, also called a page search range or page search condition. In some embodiments, the preset identity information, preset crawler conditions, and preset crawler duration can all be configured through the front-end page of the risk assessment device before risk assessment.

[0046] In step S30, one or more specified text contents in the target webpage corresponding to each target URL are parsed to obtain one or more target text vectors. One specified text content corresponds to one target text vector. For example, according to the first-in-first-out rule of the target URL queue determined above, one or more specified text contents in the target webpage corresponding to each target URL are parsed in turn to obtain one or more target text vectors.

[0047] In some embodiments, a tag tree of a target webpage corresponding to each target URL is obtained, and the text content (specified text content) of the leaf nodes in the tag tree is parsed to obtain one or more target text vectors. Figure 2 shown.

[0048] Figure 2 is a schematic diagram showing a tag tree of a certain target URL according to some embodiments of the present disclosure.

[0049] like Figure 2As shown in the figure, the tag tree of a target URL includes html nodes, head nodes, body nodes, title nodes, div nodes, null nodes, p nodes, and a nodes. The null node means that there is no node at the same level as the div node under the body node. The title node, p node, and a node are all leaf nodes, and the sub-tables include title information, paragraph information, and anchor text information. The relationship and meaning of each node are the general meanings of the front-end tags, which will not be repeated here. Figure 2 By parsing the title information, paragraph information, and anchor text information of the title node, p node, and a node, the target text vector can be obtained.

[0050] return Figure 1 , in step S50, for each target URL, the privacy type of each target text vector is determined. For example, a preset regular expression is used to perform regular matching on each target text vector. For each target text vector, if the match fails, the privacy type of the target text vector is determined to be the first privacy type. If the match is successful, the privacy type of the target text vector is determined to be the second privacy type. The first privacy type is a privacy type that does not have a fixed regular format (that is, the target text vector that cannot be identified using a fixed data format belongs to the first privacy type), for example, including multiple first subtypes such as name and address. The second privacy type is a privacy type that has a fixed regular format (that is, the target text vector that can be identified using a fixed data format belongs to the second privacy type), for example, including multiple second subtypes such as mobile phone number, ID number, bank card number, etc.

[0051] In step S70, for each target URL, the risk value of one or more specified text contents causing page privacy leakage in the target system is determined according to the privacy type of one or more target text vectors. The risk value is positively correlated with the possibility of causing page privacy leakage in the target system.

[0052] In some embodiments, the one or more designated text contents are multiple, and the one or more target text vectors are also multiple. The multiple target text vectors include at least one first target text vector whose privacy type is a first privacy type.

[0053] For example, by Figure 3 The above step S70 is implemented in the manner shown.

[0054] Figure 3 The flowchart is a flowchart showing how to determine the risk value of page privacy leakage in a target system caused by one or more specified text contents according to some embodiments of the present disclosure.

[0055] like Figure 3As shown, determining the risk value of page privacy leakage in the target system caused by one or more specified text contents includes steps S710 to S720.

[0056] In step S710, the similarity between the first target text vector space and the reference privacy feature vector space is determined according to the first subtype of at least one first target text vector. The first target text vector space includes at least one first target text vector, and the reference privacy feature vector space includes one or more reference privacy feature vectors.

[0057] In some embodiments, the reference privacy feature vector is obtained as follows.

[0058] First, crawl multiple reference text contents of a specified web page. Then, use the Gensim model to obtain the first subtype of each reference text content and the relevance of the reference text content to the first subtype. Finally, determine the text vectors of one or more reference text contents whose relevance is greater than the relevance threshold as reference privacy feature vectors. For example, the risk assessment method includes the step of obtaining a reference privacy feature vector.

[0059] In some embodiments, according to the "GB35273-2020 Personal Information Protection Specification", multiple reference text contents of a specified web page are crawled using the Skip-gram model in the word2vec (word to vector) algorithm, combined with the sensitivity level defined in the "Personal Information Protection Specification", to construct a reference privacy feature vector space.

[0060] In some embodiments, the Figure 4 Step S710 is implemented in the manner shown.

[0061] Figure 4 is a flowchart illustrating determining the similarity between a first target text vector space and a reference privacy feature vector space according to a first subtype of at least one first target text vector according to some embodiments of the present disclosure.

[0062] like Figure 4 As shown, according to the first subtype of at least one first target text vector, determining the similarity between the first target text vector space and the reference privacy feature vector space includes steps S711 and S712.

[0063] In step S711, a type weight value corresponding to a first subtype of at least one first target text vector is obtained. In some embodiments, the type weight value corresponding to a first subtype of at least one first target text vector may be obtained based on a correspondence between the first subtype and the type weight value. For example, the type weight value corresponding to the first subtype of the i-th first target text vector is w i .

[0064] In step S712, the similarity between the first target text vector space and the reference privacy feature vector space is determined based on the frequency of at least one first target text vector appearing in the target web page and at least one type weight value. For example, the frequency of the i-th first target text vector appearing in the target web page is f i .

[0065] For example, step S712 is implemented in the following manner.

[0066] First, for each first target text vector, a sensitivity weight value of each first target text vector is determined according to the frequency of occurrence of at least one first target text vector in the target web page and at least one type weight value. The sensitivity weight value is positively correlated with the privacy sensitivity of each first target text vector.

[0067] In some embodiments, at least one first target text vector is multiple. For each first target text vector, the sensitivity weight value of each first target text vector is determined based on the square sum of the product of the frequency of each first target text vector and the type weight value, and the product of the frequency of each first target text vector and the type weight value. For example, the sensitivity weight value of the i-th first target text vector is n i , Wherein, n1 is the total number of the first target text vectors, and i and n1 are both positive integers.

[0068] Then, according to the sensitivity weight value of at least one first target text vector, the similarity between the first target text vector space and the reference privacy feature vector space is determined.

[0069] In some embodiments, there are multiple at least one first target text vector and multiple one or more reference privacy feature vectors, and the weighted sum of the multiple first target text vectors is calculated using the sensitivity weight values ​​of the multiple first target text vectors; and the similarity between the first target text vector space and the reference privacy feature vector space is determined based on the weighted sum and the sum of the multiple reference privacy feature vectors.

[0070] In some embodiments, the similarity between the first target text vector space and the reference privacy feature vector space is determined based on the product of the weighted sum and the sum of multiple reference privacy feature vectors, the square sum of multiple reference privacy feature vectors, and the square sum of the product of each first target text vector and the corresponding sensitivity weight value. For example, the similarity calculation formula is: Wherein, j is the identifier of the reference privacy feature vector, n2 is the total number of reference privacy feature vectors, and both j and n2 are positive integers.

[0071] return Figure 3In step S720, based on the similarity, a risk value of page privacy leakage in the target system caused by the designated text content corresponding to at least one first target text vector is determined. The risk value is positively correlated with the similarity.

[0072] In some embodiments, the similarity may be directly determined as a risk value.

[0073] In other embodiments, the risk value of page privacy leakage in the target system caused by the specified text content corresponding to each first target text vector can be determined based on the similarity and the current system authority value. The current system authority value is positively correlated with the size of the authority to access the target system. For example, the current system authority value is pre-set by the risk assessor in the front-end page of the risk assessment device. The risk value is the ratio of the similarity to the current system authority value.

[0074] For example, the risk value of page privacy leakage in the target system caused by the designated text content corresponding to the at least one first target text vector is determined according to the similarity and the system authority value in the following manner.

[0075] First, determine the target authority value based on the current system authority value and the system authority discrete rate. The target authority value is positively correlated with the current system authority value and the system authority discrete rate, respectively. The system authority discrete rate is used to expand the distinction between different system authority values. For example, the current system authority value is a positive number a, and the system authority discrete rate is a positive integer d. The system authority discrete rate can avoid the problem that the difference between different system authority values ​​is too small, resulting in unclear changes in risk values, and can improve the accuracy of risk assessment. For example, the target authority value is 1 / a d .

[0076] Then, based on the similarity and the target permission value, the risk value of the specified text content corresponding to at least one first target text vector causing page privacy leakage in the target system is determined. The risk value is negatively correlated with the target permission value. For example, the calculation formula of the risk value R is:

[0077] In some embodiments, the plurality of target text vectors further include at least one second target text vector of a second privacy type. Determining the risk value of page privacy leakage in the target system caused by one or more specified text contents may be achieved in the following manner.

[0078] First, the target permission value is determined according to the current system permission value and the system permission discrete rate.

[0079] Then, according to the target permission value, the risk value of the specified text content corresponding to at least one second target text vector causing the target system to leak page privacy is determined. That is, in this case, the calculation formula of the risk value R is:

[0080] In some embodiments, the correspondence between the target URL, the specified text content and the risk value can be displayed to the risk assessor so that the risk assessor can conduct manual investigation based on the acquired correspondence or optimize and iterate the preset current system permission value, system permission discrete rate, type weight value and other configuration parameters.

[0081] In the above embodiment, the target text vector of the target URL is automatically acquired and parsed, and according to the privacy type of the target text vector, the risk value of page privacy leakage in the target system caused by the specified text content is determined, and the risk assessment process of page privacy leakage is quantified and automated, solving the problems of low page coverage, low inspection efficiency and strong subjectivity in risk judgment caused by manual inspection, thereby improving the efficiency and accuracy of risk assessment.

[0082] Figure 5 is a block diagram illustrating a risk assessment system according to some embodiments of the present disclosure.

[0083] like Figure 5 As shown, the risk assessment system 5 includes a risk assessment device 51 and a target system 52. The target system 52 includes a target front-end page, and the target front-end page is used for risk assessment.

[0084] The risk assessment device 51 includes an interactive interface 511 , a policy storage module 512 , a subject search engine 513 , a data processing module 514 , a database 515 , a risk assessment module 516 and a vector space construction module 517 .

[0085] The interactive interface 511 is configured to interface the configuration information filled in by the risk assessor on the front-end configuration page of the risk assessment device 51. The interactive interface 511 includes a policy configuration interface 5111.

[0086] In some embodiments, the risk assessor fills in crawler strategies such as preset identity information, one or more preset seed URLs, preset crawler conditions, and preset crawler duration on the front-end configuration page of the risk assessment device 51. Furthermore, the front-end configuration page sends the crawler strategy configured by the risk assessor to the strategy storage module 512 via the strategy configuration interface 5111 of the interactive interface 511.

[0087] The subject search engine 513 is configured to obtain the crawler strategy from the strategy storage module 512, and obtain multiple target URLs corresponding to the target system according to the crawler strategy.

[0088] For example, the subject search engine 513 is configured to use the preset identity information to log in to the target system 52 to crawl one or more web pages related to one or more preset seed URLs; and determine the target URL queue according to the relevance of the crawled one or more web pages to the one or more preset seed URLs. The target URL queue includes multiple target URLs arranged in descending order according to the relevance, and the multiple target URLs are at least a part of the URLs of one or more web pages.

[0089] The subject search engine 513 is also configured as a first-in-first-out rule for the target URL queue, and sequentially parses one or more specified text contents in the target web page corresponding to each target URL to obtain one or more target text vectors.

[0090] In some embodiments, the subject search engine 513 sends the parsed one or more target text vectors of each target URL to the data processing module 514. The data processing module 514 is configured to perform data cleaning and data assembly on the one or more target text vectors of each target URL.

[0091] For example, data cleaning includes deleting outliers in one or more target text vectors of each target URL, and data assembly includes structurally assembling one or more target text vectors of each target URL after data cleaning and storing them in database 515. Database 515 is, for example, a database MongoDB based on distributed file storage.

[0092] The risk assessment module 516 is configured to obtain one or more target text vectors for each target URL in the database 515 , and determine the privacy type of each target text vector using the included matching unit 5161 .

[0093] For example, the matching unit 5161 uses a preset regular expression to perform regular matching on each target text vector. For each target text vector, if the match fails, the privacy type of the target text vector is determined to be the first privacy type. If the match is successful, the privacy type of the target text vector is determined to be the second privacy type. The first privacy type is a privacy type without a fixed regular format, such as a name, address, and other first subtypes. The second privacy type is a privacy type with a fixed regular format, such as a mobile phone number, ID card number, bank card number, and other second subtypes.

[0094] The vector space construction module 517 is configured to crawl multiple reference text contents of a specified web page. Then, using the Gensim model, the first subtype of each reference text content and the relevance of the reference text content to the first subtype are obtained. Finally, the text vectors of one or more reference text contents whose relevance is greater than the relevance threshold are determined as reference privacy feature vectors. The one or more reference privacy feature vectors determined by the vector space construction module 517 constitute a reference privacy feature vector space.

[0095] The risk assessment module 516 also includes a sensitivity analysis unit 5162 and a similarity analysis unit 5163 .

[0096] The sensitivity analysis unit 5162 is configured to determine the sensitivity weight value of each first target text vector for each first target text vector according to the frequency of occurrence of at least one first target text vector in the target web page and at least one type weight value. The sensitivity weight value is positively correlated with the privacy sensitivity of each first target text vector. When the first target text vector is one or more specified text contents, the target text vector whose privacy type is the first privacy type among the multiple target text vectors. The specific implementation process may be the same or similar to the steps disclosed in the aforementioned embodiments, and will not be repeated here.

[0097] The similarity analysis unit 5163 is configured to determine the similarity between the first target text vector space and the reference privacy feature vector space according to the sensitivity weight value of at least one first target text vector obtained from the sensitivity analysis unit 5162. The first target text vector space includes at least one first target text vector. The specific implementation process may be the same or similar to the steps disclosed in the aforementioned embodiment, and will not be repeated here.

[0098] The interactive interface 511 also includes a permission configuration interface 5112. The risk assessor fills in the current system permissions and permission discrete rate on the front-end configuration page of the risk assessment device 51, and the front-end configuration page sends the current system permissions and permission discrete rate configured by the risk assessor to the permission analysis unit 5164 of the risk assessment module 516 through the permission configuration interface 5112.

[0099] The authority analysis unit 5164 is configured to determine the target authority value according to the current system authority and the authority discrete rate. The target authority value is positively correlated with the current system authority value and the system authority discrete rate, respectively. The system authority discrete rate is used to expand the distinction between different system authority values. For example, the current system authority value is a, and the system authority discrete rate is d. The system authority discrete rate can avoid the problem that the difference between different system authority values ​​is too small, resulting in unclear changes in risk values, and can improve the accuracy of risk assessment. For example, the target authority value is 1 / a d .

[0100] The risk assessment module 516 also includes a risk value determination unit 5165. The risk value determination unit 5165 is configured to determine, for each target URL, a risk value of one or more specified text contents causing page privacy leakage in the target system according to the privacy type of one or more target text vectors.

[0101] For example, for the first target text vector, the risk value of the target system's page privacy leakage caused by the specified text content corresponding to at least one first target text vector is determined based on the similarity obtained from the similarity analysis unit 5163 and the target permission value obtained from the permission analysis unit 5164. The risk value is positively correlated with the similarity and negatively correlated with the target permission value.

[0102] For another example, for the second target text vector, the risk value of the specified text content corresponding to at least one second target text vector causing the target system to have page privacy leakage is determined according to the target permission value obtained from the permission analysis unit 5164. When the second target text vector is one or more specified text contents, the target text vector of the privacy type of the second privacy type among the multiple target text vectors.

[0103] The interactive interface 511 also includes a data display interface 5113 .

[0104] After determining the risk value, the risk value determination unit 5165 displays the target URL, the designated text content corresponding to the first target text vector or the second target text vector, and the corresponding risk value on the front-end display page of the risk assessment device 51 through the data display interface 5113 .

[0105] Figure 6 is a block diagram illustrating a risk assessment apparatus according to some embodiments of the present disclosure.

[0106] like Figure 6 As shown, the risk assessment device 61 includes an acquisition module 611, an analysis module 612, a first determination module 613 and a second determination module 614. The risk assessment device 61 is, for example, Figure 5 The structure and function of the risk assessment device 51 are the same or similar.

[0107] The acquisition module 611 is configured to acquire multiple target uniform resource locators URLs corresponding to the target system, for example, by executing Figure 1 Step S10 is shown.

[0108] The parsing module 612 is configured to parse one or more specified text contents in the target webpage corresponding to each target URL to obtain one or more target text vectors, where one specified text content corresponds to one target text vector. For example, Figure 1Step S30 is shown.

[0109] The first determination module 613 is configured to determine the privacy type of each target text vector for each target URL, for example, by performing the following steps: Figure 1 Step S50 is shown.

[0110] The second determination module 614 is configured to determine, for each target URL, a risk value of one or more specified text contents causing page privacy leakage in the target system according to the privacy type of one or more target text vectors, for example, by performing the following steps: Figure 1 Step S70 is shown.

[0111] Figure 7 is a block diagram showing a risk assessment device according to some other embodiments of the present disclosure.

[0112] like Figure 7 As shown, the risk assessment device 71 includes a memory 711; and a processor 712 coupled to the memory 711. The memory 711 is used to store instructions for executing the corresponding embodiments of the risk assessment method. The processor 712 is configured to execute the risk assessment method in any of the embodiments of the present disclosure based on the instructions stored in the memory 711. The structure and function of the risk assessment device 71 are similar to Figure 5 Risk assessment device 51, Figure 6 The risk assessment device 61 is the same as or similar to the risk assessment device 61.

[0113] Figure 8 is a block diagram illustrating a computer system for implementing some embodiments of the present disclosure.

[0114] like Figure 8 As shown, the computer system 80 may be embodied in the form of a general-purpose computing device. The computer system 80 includes a memory 810, a processor 820, and a bus 800 that connects the various system components.

[0115] The memory 810 may include, for example, a system memory, a non-volatile storage medium, etc. The system memory may store, for example, an operating system, an application program, a boot loader, and other programs. The system memory may include a volatile storage medium, such as a random access memory (RAM) and / or a cache memory. The non-volatile storage medium may store, for example, instructions for executing at least one corresponding embodiment of the risk assessment method. The non-volatile storage medium may include, but is not limited to, a disk memory, an optical memory, a flash memory, etc.

[0116] The processor 820 can be implemented by a general processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistors, etc. Discrete hardware components. Accordingly, each module such as the judgment module and the determination module can be implemented by a central processing unit (CPU) running instructions in a memory that execute corresponding steps, or can be implemented by a dedicated circuit that executes corresponding steps.

[0117] The bus 800 may use any of a variety of bus architectures, including, but not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MCA) bus, and a Peripheral Component Interconnect (PCI) bus.

[0118] The computer system 80 may also include an input / output interface 830, a network interface 840, a storage interface 850, etc. These interfaces 830, 840, 850, the memory 810, and the processor 820 may be connected via a bus 800. The input / output interface 830 may provide a connection interface for input / output devices such as a display, a mouse, and a keyboard. The network interface 840 may provide a connection interface for various networked devices. The storage interface 850 may provide a connection interface for external storage devices such as a floppy disk, a USB flash drive, and an SD card.

[0119] Here, various aspects of the present disclosure are described with reference to flowcharts and / or block diagrams of methods, devices, and computer program products according to embodiments of the present disclosure. It should be understood that each frame of the flowchart and / or block diagram and the combination of frames can be implemented by computer-readable program instructions.

[0120] These computer-readable program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable device to produce a machine, so that the processor executes the instructions to produce means for implementing the functions specified in one or more blocks in the flowchart and / or block diagram.

[0121] These computer-readable program instructions may also be stored in a computer-readable memory, which cause the computer to work in a specific manner to produce an article of manufacture, including instructions for implementing the functions specified in one or more blocks in the flowchart and / or block diagram.

[0122] The present disclosure can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects.

[0123] The risk assessment method and device, and computer storable medium in the above embodiments can improve the efficiency and accuracy of risk assessment.

[0124] So far, the risk assessment method and device, and computer storable medium according to the present disclosure have been described in detail. In order to avoid obscuring the concept of the present disclosure, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solution disclosed herein.

Claims

1. A risk assessment method comprising: Obtain multiple target uniform resource locator URLs corresponding to the target system; Parsing multiple specified text contents in the target webpage corresponding to each target URL to obtain multiple target text vectors, where one specified text content corresponds to one target text vector; For each target URL, determining the privacy type of each target text vector includes: using a regular expression to perform regular matching on each target text vector; if the matching fails, determining the privacy type of the target text vector as a first privacy type, otherwise, determining the privacy type of the target text vector as a second privacy type, wherein the first privacy type includes a plurality of first subtypes; For each of the target URLs, determining, according to the privacy types of the multiple target text vectors, the risk value of the target system causing the page privacy leakage due to the multiple specified text contents, including: in a case where the multiple target text vectors include at least one first target text vector of a first privacy type, determining, according to a first subtype of the at least one first target text vector, the similarity between a first target text vector space and a reference privacy feature vector space, the first target text vector space including the at least one first target text vector, and the reference privacy feature vector space including one or more reference privacy feature vectors; determining, according to the similarity, the risk value of the target system causing the page privacy leakage due to the specified text content corresponding to the at least one first target text vector, the risk value being positively correlated with the similarity.

2. The risk assessment method according to claim 1, wherein: Determining the similarity between the first target text vector space and the reference privacy feature vector space according to the first subtype of the at least one first target text vector comprises: Obtaining a type weight value corresponding to a first subtype of the at least one first target text vector; The similarity between the first target text vector space and the reference privacy feature vector space is determined according to the frequency of occurrence of the at least one first target text vector in the target web page and at least one type weight value.

3. The risk assessment method according to claim 2, wherein: Determining the similarity between the first target text vector space and the reference privacy feature vector space according to the frequency of occurrence of the at least one first target text vector in the target webpage and at least one type weight value includes: For each first target text vector, determining a sensitivity weight value of each first target text vector according to the frequency of occurrence of the at least one first target text vector in the target webpage and the at least one type weight value, wherein the sensitivity weight value is positively correlated with the privacy sensitivity of each first target text vector; According to the sensitivity weight value of the at least one first target text vector, the similarity between the first target text vector space and the reference privacy feature vector space is determined.

4. The risk assessment method according to claim 3, wherein: There are multiple first target text vectors, and determining the sensitivity weight value of each first target text vector includes: For each first target text vector, the sensitivity weight value of each first target text vector is determined based on the square sum of the products of the frequency and the type weight value of each first target text vector and the product of the frequency and the type weight value of each first target text vector.

5. The risk assessment method according to claim 3, wherein: There are multiple at least one first target text vector, there are multiple one or more reference privacy feature vectors, and according to the sensitivity weight value of the at least one first target text vector, determining the similarity between the first target text vector space and the reference privacy feature vector space comprises: Calculating a weighted sum of the plurality of first target text vectors using the sensitivity weight values ​​of the plurality of first target text vectors; The similarity between the first target text vector space and the reference privacy feature vector space is determined according to the weighted sum and the sum of multiple reference privacy feature vectors.

6. The risk assessment method according to claim 5, wherein: Determining the similarity between the first target text vector space and the reference privacy feature vector space according to the weighted sum and the sum of the multiple reference privacy feature vectors includes: The similarity between the first target text vector space and the reference privacy feature vector space is determined based on the product of the weighted sum and the sum of multiple reference privacy feature vectors, the sum of the squares of the multiple reference privacy feature vectors, and the sum of the squares of the products of each first target text vector and the corresponding sensitivity weight value.

7. The risk assessment method according to claim 1, wherein: Determining, according to the similarity, a risk value of page privacy leakage in the target system caused by the specified text content corresponding to each of the first target text vectors includes: A risk value of page privacy leakage in the target system caused by the specified text content corresponding to each first target text vector is determined based on the similarity and the current system permission value, wherein the current system permission value is positively correlated with the permission size for accessing the target system.

8. The risk assessment method according to claim 7, wherein: Determining, according to the similarity and the system authority value, a risk value of page privacy leakage in the target system caused by the specified text content corresponding to the at least one first target text vector includes: Determine a target authority value according to the current system authority value and the system authority discrete rate, wherein the target authority value is positively correlated with the current system authority value and the system authority discrete rate, respectively, and the system authority discrete rate is used to expand the distinction between different system authority values; A risk value of page privacy leakage in the target system caused by designated text content corresponding to the at least one first target text vector is determined according to the similarity and the target authority value, wherein the risk value is negatively correlated with the target authority value.

9. The risk assessment method according to claim 1, further comprising: Crawl multiple reference text contents of a specified web page; Using the Gensim model, obtaining the first subtype of each reference text content and the relevance between the reference text content and the first subtype; Determine one or more text vectors of reference text content whose relevance is greater than a relevance threshold as the reference privacy feature vector.

10. The risk assessment method according to claim 1, wherein: The multiple target text vectors include at least one second target text vector of a second privacy type, and determining the risk value of page privacy leakage of the target system caused by the multiple specified text contents includes: Determine a target permission value according to a current system permission value and a system permission discrete rate, wherein the current system permission value is positively correlated with the permission size for accessing the target system, and the target permission value is positively correlated with the current system permission value and the system permission discrete rate, respectively, and the system permission discrete rate is used to expand the distinction between different system permission values; A risk value of page privacy leakage in the target system caused by designated text content corresponding to the at least one second target text vector is determined according to the target authority value, wherein the risk value is negatively correlated with the target authority value.

11. The risk assessment method according to claim 1, wherein: Obtaining multiple target uniform resource locator URLs corresponding to the target system includes: Using the preset identity information, crawl one or more web pages associated with one or more preset seed URLs; According to the relevance of the one or more web pages to the one or more preset seed URLs, a target URL queue is determined, wherein the target URL queue includes a plurality of target URLs arranged in descending order according to the relevance, and the plurality of target URLs are at least a part of the URLs of the one or more web pages.

12. The risk assessment method according to claim 11, wherein: Parsing multiple specified text contents in the target webpage corresponding to each target URL to obtain multiple target text vectors includes: According to the first-in-first-out rule of the target URL queue, multiple specified text contents in the target webpage corresponding to each target URL are parsed in turn to obtain multiple target text vectors.

13. The risk assessment method according to claim 11, wherein: Using the preset identity information, crawling one or more web pages associated with one or more preset seed URLs includes: Using the preset identity information, crawl one or more web pages related to one or more preset seed URLs and meeting the preset crawler conditions until the crawling time is equal to the preset crawler time.

14. A risk assessment device, comprising: An acquisition module is configured to acquire a plurality of target uniform resource locators URLs corresponding to the target system; A parsing module is configured to parse multiple specified text contents in a target webpage corresponding to each target URL to obtain multiple target text vectors, where one specified text content corresponds to one target text vector; A first determination module is configured to determine the privacy type of each target text vector for each target URL, including: using a regular expression to perform regular matching on each target text vector; if the matching fails, determining the privacy type of the target text vector as a first privacy type, otherwise, determining the privacy type of the target text vector as a second privacy type, wherein the first privacy type includes a plurality of first subtypes; The second determination module is configured to determine, for each target URL, the risk value of the target system being caused by the multiple specified text contents causing page privacy leakage according to the privacy types of the multiple target text vectors, including: when the multiple target text vectors include at least one first target text vector of a first privacy type, determining the similarity between the first target text vector space and the reference privacy feature vector space according to the first subtype of the at least one first target text vector, the first target text vector space including the at least one first target text vector, and the reference privacy feature vector space including one or more reference privacy feature vectors; and determining, according to the similarity, the risk value of the target system being caused by the specified text content corresponding to the at least one first target text vector, the risk value being positively correlated with the similarity.

15. A risk assessment device, comprising: Memory; as well as A processor coupled to the memory, wherein the processor is configured to execute the risk assessment method according to any one of claims 1 to 13 based on instructions stored in the memory.

16. A computer storable medium having computer program instructions stored thereon, which, when executed by a processor, implement the risk assessment method according to any one of claims 1 to 13.

Images