A data transaction compliance detection system and method

Abstract

The present application provides a data transaction compliance detection system and method, the system comprising: a data acquisition module for collecting data to be detected including data directory information, transaction user information, data source information and data authorization information; a data evaluation module for retrieving each piece of data to be detected, performing semantic analysis on the data to be detected, generating one or more keywords, matching the keywords with a sensitive word library, and outputting a yes or no matching result; a data analysis module for retrieving a data analysis model, inputting the data to be detected into the data analysis model when the matching result is no, analyzing the sensitive features of the data to be detected, and outputting the analysis result; a data reporting module for generating an evaluation report based on the matching result and the analysis result. The present application improves the efficiency and accuracy of data transaction compliance detection by automatically detecting data transaction compliance.

Description

Technical Field

[0001] The present application seeks to protect a data evaluation technology, and more particularly, relates to a data transaction compliance detection system. The present application also relates to a data transaction compliance detection method. Background Art

[0002] As a new type of asset, data assets have played an important role in corporate operations. With the development of electronic information technology, data assets will be more important in many fields such as enterprises and society.

[0003] However, most data assets are collected based on a large amount of user personal information. This data is sensitive and important, and is subject to legal usage regulations including the Data Security Law. However, it is still impossible to fully conduct data transaction compliance audits through manual review, which leads to data transaction risks and personal user data security issues.

[0004] At present, there is no technical means to detect the compliance of the data transaction process. Summary of the invention

[0005] In order to solve the technical problems mentioned in the above background technology, the present application provides a data transaction compliance detection system. The present application also relates to a data transaction compliance detection method.

[0006] The present application provides a data transaction compliance detection system, including:

[0007] The data collection module is used to collect the data to be tested, including data directory information, transaction user information, data source information and data authorization information;

[0008] The data evaluation module is used to retrieve each piece of data to be detected, perform semantic analysis on the data to be detected, generate one or more keywords, match the keywords with the sensitive word library, and output a yes or no matching result;

[0009] A data analysis module, used to call a data analysis model, and when the matching result is negative, input the data to be detected into the data analysis model, analyze the sensitive features of the data to be detected, and output the analysis result;

[0010] A data reporting module is used to generate an evaluation report based on the matching results and the analysis results.

[0011] Optionally, matching the keyword with a sensitive word library includes:

[0012] Obtain the keyword, and obtain the context of the keyword;

[0013] Retrieving the sensitive word library, comparing the sensitive word library with the keyword and the context of the keyword, and obtaining the number of matching words;

[0014] Calculate the ratio of the number of matching words to the keyword and the context of the keyword;

[0015] According to a preset threshold of the ratio, it is determined whether the keyword matches the sensitive word.

[0016] Optionally, the sensitive word library is stored in a read-only memory.

[0017] Optionally, also include:

[0018] The age judgment module is used to obtain multiple data information and cross-judge the multiple data information, including: identifying the identity attribute information of each data information respectively to form an identity data set; and judging the age range according to the identity data set.

[0019] Optionally, also include:

[0020] The information recommendation module is used to match the recommended information from the database according to the evaluation report and write the recommended information into the evaluation report.

[0021] This application also provides a data transaction compliance detection method, including:

[0022] Collect the data to be tested, including data directory information, transaction user information, data source information, and data authorization information;

[0023] Retrieve each piece of data to be tested, perform semantic analysis on the data to be tested, generate one or more keywords, match the keywords with the sensitive word library, and output a yes or no match result;

[0024] Retrieving a data analysis model, and when the matching result is negative, inputting the data to be detected into the data analysis model, analyzing the sensitive features of the data to be detected, and outputting the analysis result;

[0025] An evaluation report is generated according to the matching result and the analysis result.

[0026] Optionally, matching the keyword with a sensitive word library includes:

[0027] Obtain the keyword, and obtain the context of the keyword;

[0028] Retrieving the sensitive word library, comparing the sensitive word library with the keyword and the context of the keyword, and obtaining the number of matching words;

[0029] Calculate the ratio of the number of matching words to the keyword and the context of the keyword;

[0030] According to a preset threshold of the ratio, it is determined whether the keyword matches the sensitive word.

[0031] Optionally, the sensitive word library is stored in a read-only memory.

[0032] Optionally, also include:

[0033] Acquire multiple data information and cross-judge the multiple data information, including: respectively identifying identity attribute information of each data information to form an identity data set; and judging the age range according to the identity data set.

[0034] Optionally, also include:

[0035] According to the evaluation report, recommendation information is matched from a database, and the recommendation information is written into the evaluation report.

[0036] The advantages of this application compared to the prior art are:

[0037] The present application provides a data transaction compliance detection system, including: a data acquisition module, which is used to collect data to be detected, including data directory information, transaction user information, data source information and data authorization information; a data evaluation module, which is used to retrieve each piece of data to be detected, perform semantic analysis on the data to be detected, generate one or more keywords, match the keywords with a sensitive word library, and output a matching result of yes or no; a data analysis module, which is used to retrieve a data analysis model, and when the matching result is no, input the data to be detected into the data analysis model, analyze the sensitive features of the data to be detected, and output the analysis result; a data reporting module, which is used to generate an evaluation report based on the matching result and the analysis result. The present application improves the efficiency and accuracy of data transaction compliance detection by automatically detecting data transaction compliance. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a schematic diagram of the data transaction compliance detection system in this application.

[0039] Figure 2 is a flowchart of data evaluation in this application.

[0040] Figure 3 This is the data transaction compliance detection flow chart in this application. DETAILED DESCRIPTION

[0041] The following contents are all examples of specific implementation processes provided for describing in detail the technical solutions to be protected by this application. However, this application may also be implemented in other ways different from the descriptions herein. Persons skilled in the art may, under the guidance of the concepts of this application, adopt different technical means to implement this application. Therefore, this application is not limited to the specific embodiments below.

[0042] The present application provides a data transaction compliance detection system, including: a data acquisition module, which is used to collect data to be detected, including data directory information, transaction user information, data source information and data authorization information; a data evaluation module, which is used to retrieve each piece of data to be detected, perform semantic analysis on the data to be detected, generate one or more keywords, match the keywords with a sensitive word library, and output a matching result of yes or no; a data analysis module, which is used to retrieve a data analysis model, and when the matching result is no, input the data to be detected into the data analysis model, analyze the sensitive features of the data to be detected, and output the analysis result; a data reporting module, which is used to generate an evaluation report based on the matching result and the analysis result. The present application improves the efficiency and accuracy of data transaction compliance detection by automatically detecting data transaction compliance.

[0043] Figure 1 This is a schematic diagram of the data transaction compliance detection system in this application.

[0044] Please refer to Figure 1 As shown, the data collection module 101 is used to collect the data to be detected including data directory information, transaction user information, data source information and data authorization information.

[0045] The data refers to data assets, which can be traded. Compliance testing is required during the transaction process to determine the compliance of the data and data transactions. The compliance testing of the data is to determine whether the source, processing and other data-related operations are compliant in accordance with the provisions of relevant national laws and regulations.

[0046] During the data transaction process, all the traded data can generate a data set, which is stored in the network system in a fixed hierarchical relationship. Alternatively, it can be understood that the data in the data set have an association relationship, such as different types of data of the same object, different objects of the same type of data, etc.

[0047] The data acquisition module 101 of the present application performs scanning based on the data set, obtains the directory of the data set, and generates data directory information. Specifically, the directory information contains the association relationship of the data in the data set, and the association relationship is readable. A preferred method is that the association relationship is marked with different tags or special fields.

[0048] After analyzing and completing the collection of the data directory information, the data collection module 101 collects the user information, data source information and data authorization information of the two parties of the data set transaction in sequence. Preferably, the data source information is determined. If the source is the transaction party, the data authorization information may no longer be determined, because if the data source is the transaction party, the transaction of the data set is invalid without authorization by others.

[0049] From then on, the data collection module 101 described in the present application obtains the data to be detected including data directory information, transaction user information, data source information and data authorization information.

[0050] Please refer to Figure 1 As shown, the data evaluation module 102 is used to retrieve each piece of data to be detected, perform semantic analysis on the data to be detected, generate one or more keywords, match the keywords with a sensitive word library, and output a yes or no matching result.

[0051] After the data to be detected is collected by the data collection module 101, the data evaluation module 102 obtains the data to be detected from the data collection module 101 and performs further processing. The sensitive word library is stored in a read-only memory.

[0052] Figure 2 is a flowchart of data evaluation in this application.

[0053] Please refer to Figure 2 As shown, S201 obtains the keyword and obtains the context of the keyword.

[0054] The data evaluation module 102 sequentially acquires the acquired data to be detected, acquiring one piece of the data to be detected each time, and then analyzes the data to be detected to generate one or more keywords.

[0055] The semantic analysis specifically refers to semantic analysis at the vocabulary level, which uses a natural language recognition method to segment the data to be detected at the vocabulary level to form one or more keywords.

[0056] Further, the context of the keyword is obtained, and the context is a complete sentence or a complete paragraph in which the keyword is located. In the specific implementation process, the context can be set according to the specific situation to form contexts of different ranges.

[0057] S202 retrieves the sensitive word library, compares the sensitive word library with the keyword and the context of the keyword, and obtains the number of matching words.

[0058] The number of matching words refers to the number of matching keywords. Specifically, firstly, a sensitive word library is retrieved, and the sensitive words in the sensitive word library are compared to obtain a comparison result between the keyword and the sensitive word.

[0059] It should be noted that, in the present application, if the keyword is compared with the sensitive word library, and the comparison result is that the keyword is a sensitive word, then no further comparison of the context with the sensitive word library is required.

[0060] If not, the context is parsed and identified, divided into a plurality of comparison words through natural language recognition, and the comparison words are compared with the sensitive word library to determine the number of matching words.

[0061] S203 calculates the ratio of the number of matching words to the keyword and the context of the keyword, and the specific method is to divide the number of matching words by the number of comparison words.

[0062] S104 determines whether the keyword matches the sensitive word according to a preset threshold of the ratio.

[0063] In this application, when the ratio reaches 30%, it can be considered that the keyword and the sensitive word have a matching relationship. Of course, further determination is required, including further determination by manual identification. If it is determined that the keyword is a sensitive word, the keyword is labeled and added to the sensitive word library.

[0064] Please refer to Figure 1 As shown, the data analysis module 103 is used to call the data analysis model, and when the matching result is no, input the data to be detected into the data analysis model, analyze the sensitive features of the data to be detected, and output the analysis result.

[0065] The data analysis model includes:

[0066] A data input unit, used to input all keywords of the data to be detected;

[0067] The assignment unit is used to perform semantic assignment based on the keywords. The semantic assignment refers to parsing the semantics of the keywords and labeling the semantics of the keywords. The label is the assignment.

[0068] The combination identification unit performs sensitive semantic combination based on the assignments, including randomly extracting a plurality of the assignments, randomly combining the assignments, and performing sensitive semantic identification on the combined assignments.

[0069] Specifically, assuming that the data to be detected includes five keywords: A, B, C, D, and E, the five keywords are input at this time.

[0070] At this time, the keywords are identified, semantically parsed and assigned values. For example, keyword A is "primary school", and the value assigned is a, which is a label. Keyword B is a phrase "class", and the value assigned is b, which is a label.

[0071] Based on the above assignments, the assignments of keyword A and keyword B are combined to generate the assignment field "a, b" for sensitive feature matching.

[0072] The matching has a matching table, which randomly combines keywords of sensitive information and maps the value fields corresponding to the keywords to the sensitive information respectively.

[0073] Taking "a, b" as an example, it can be mapped to primary school students according to the assignment field. At this time, if primary school students are sensitive information, the analysis result of the data to be detected can be regarded as having sensitive characteristics.

[0074] Specifically, each piece of information to be detected described in the present application needs to be scored for whether it has the sensitive feature, including the assignment field sensitive feature record and score calculation.

[0075] The sensitive feature record refers to recording whether the value assigned to each randomly composed keyword is a sensitive feature after comparing the value assigned to each randomly composed keyword with the sensitive feature matching table.

[0076] The calculation formula is as follows:

[0077]

[0078] Wherein, P is the score, ranging from 0 to 1, Q is the number of randomly combined field assignments, M is the number of matched sensitive features, and Q is the number of randomly combined field assignments. l It is the total number of sensitive features of each sensitive information in the matching table.

[0079] When the score is greater than 0.3, it is considered that the data to be detected has sensitive features.

[0080] Please refer to Figure 1 As shown, the data reporting module 104 is used to generate an evaluation report based on the matching results and the analysis results.

[0081] The data reporting module 104 obtains the matching result and the analysis result respectively, and generates a report according to the matching result and the analysis result.

[0082] Preferably, the system further comprises an authentication module, which is used to obtain the matching result and the analysis result, and generate an authentication mark according to the matching result and the analysis result.

[0083] Preferably, the system further comprises an information recommendation module, which is used to match recommendation information from a database according to the evaluation report and write the recommendation information into the evaluation report.

[0084] The present application also provides a data transaction compliance detection method, which processes keywords to achieve compliance detection of data transactions.

[0085] Figure 3 This is the data transaction compliance detection flow chart in this application.

[0086] Please refer to Figure 3 As shown, S301 collects the data to be detected including data directory information, transaction user information, data source information and data authorization information.

[0087] The data refers to data assets, which can be traded. Compliance testing is required during the transaction process to determine the compliance of the data and data transactions. The compliance testing of the data is to determine whether the source, processing and other data-related operations are compliant in accordance with the provisions of relevant national laws and regulations.

[0088] During the data transaction process, all the traded data can generate a data set, which is stored in the network system in a fixed hierarchical relationship. Alternatively, it can be understood that the data in the data set have an association relationship, such as different types of data of the same object, different objects of the same type of data, etc.

[0089] Based on scanning the data set, a directory of the data set is obtained, and data directory information is generated. Specifically, the directory information contains associations of data in the data set, and the associations are readable. A preferred method is that the associations are marked with different tags or special fields.

[0090] After analyzing and completing the collection of data directory information, the user information, data source information and data authorization information of both parties of the data set transaction are collected in sequence. Preferably, the data source information is determined. If the source is the transaction party, the data authorization information can no longer be determined, because if the data source is the transaction party, the transaction of the data set is invalid without authorization by others.

[0091] Since then, the data to be tested including data directory information, transaction user information, data source information and data authorization information have been obtained.

[0092] Please refer to Figure 3As shown, S302 retrieves each piece of data to be detected, performs semantic analysis on the data to be detected, generates one or more keywords, matches the keywords with a sensitive word library, and outputs a yes or no matching result.

[0093] After the data to be detected is collected, the data to be detected is obtained and further processed. The sensitive word library is stored in a read-only memory.

[0094] Figure 2 is a flowchart of data evaluation in this application.

[0095] Please refer to Figure 2 As shown, S201 obtains the keyword and obtains the context of the keyword.

[0096] The data evaluation module 102 sequentially acquires the acquired data to be detected, acquiring one piece of the data to be detected each time, and then analyzes the data to be detected to generate one or more keywords.

[0097] The semantic analysis specifically refers to semantic analysis at the vocabulary level, which uses a natural language recognition method to segment the data to be detected at the vocabulary level to form one or more keywords.

[0098] Further, the context of the keyword is obtained, and the context is a complete sentence or a complete paragraph in which the keyword is located. In the specific implementation process, the context can be set according to the specific situation to form contexts of different ranges.

[0099] S202 retrieves the sensitive word library, compares the sensitive word library with the keyword and the context of the keyword, and obtains the number of matching words.

[0100] The number of matching words refers to the number of matching keywords. Specifically, firstly, a sensitive word library is retrieved, and the sensitive words in the sensitive word library are compared to obtain a comparison result between the keyword and the sensitive word.

[0101] It should be noted that, in the present application, if the keyword is compared with the sensitive word library, and the comparison result is that the keyword is a sensitive word, then no further comparison of the context with the sensitive word library is required.

[0102] If not, the context is parsed and identified, divided into a plurality of comparison words through natural language recognition, and the comparison words are compared with the sensitive word library to determine the number of matching words.

[0103] S203 calculates the ratio of the number of matching words to the keyword and the context of the keyword, and the specific method is to divide the number of matching words by the number of comparison words.

[0104] S204 determines whether the keyword matches the sensitive word according to a preset threshold of the ratio.

[0105] In this application, when the ratio reaches 30%, it can be considered that the keyword and the sensitive word have a matching relationship. Of course, further determination is required, including further determination by manual identification. If it is determined that the keyword is a sensitive word, the keyword is labeled and added to the sensitive word library.

[0106] Please refer to Figure 3 As shown, S303 retrieves a data analysis model, and when the matching result is negative, inputs the data to be detected into the data analysis model, analyzes the sensitive features of the data to be detected, and outputs the analysis result.

[0107] The data analysis model includes:

[0108] A data input unit, used to input all keywords of the data to be detected;

[0109] The assignment unit is used to perform semantic assignment based on the keywords. The semantic assignment refers to parsing the semantics of the keywords and labeling the semantics of the keywords. The label is the assignment.

[0110] The combination identification unit performs sensitive semantic combination based on the assignments, including randomly extracting a plurality of the assignments, randomly combining the assignments, and performing sensitive semantic identification on the combined assignments.

[0111] Specifically, assuming that the data to be detected includes five keywords: A, B, C, D, and E, the five keywords are input at this time.

[0112] At this time, the keywords are identified, semantically parsed and assigned values. For example, keyword A is "primary school", and the value assigned is a, which is a label. Keyword B is a phrase "class", and the value assigned is b, which is a label.

[0113] Based on the above assignments, the assignments of keyword A and keyword B are combined to generate the assignment field "a, b" for sensitive feature matching.

[0114] The matching has a matching table, which randomly combines keywords of sensitive information and maps the value fields corresponding to the keywords to the sensitive information respectively.

[0115] Taking "a, b" as an example, it can be mapped to primary school students according to the assignment field. At this time, if primary school students are sensitive information, the analysis result of the data to be detected can be regarded as having sensitive characteristics.

[0116] Specifically, each piece of information to be detected described in the present application needs to be scored for whether it has the sensitive feature, including the assignment field sensitive feature record and score calculation.

[0117] The sensitive feature record refers to recording whether the value assigned to each randomly composed keyword is a sensitive feature after comparing the value assigned to each randomly composed keyword with the sensitive feature matching table.

[0118] The calculation formula is as follows:

[0119]

[0120] Wherein, P is the score, ranging from 0 to 1, Q is the number of randomly combined field assignments, M is the number of matched sensitive features, and Q is the number of randomly combined field assignments. l It is the total number of sensitive features of each sensitive information in the matching table.

[0121] When the score is greater than 0.3, it is considered that the data to be detected has sensitive features.

[0122] Please refer to Figure 3 As shown, S304 generates an evaluation report according to the matching result and the analysis result.

[0123] The matching result and the analysis result are obtained respectively, and a report is generated according to the matching result and the analysis result.

[0124] Preferably, the method further includes obtaining the matching result and the analysis result, and generating an authentication mark according to the matching result and the analysis result.

[0125] Preferably, the method further includes matching recommendation information from a database according to the evaluation report, and writing the recommendation information into the evaluation report.

Claims

1. A data transaction compliance detection system, characterized in that: include: The data collection module is used to collect the data to be tested, including data directory information, transaction user information, data source information and data authorization information; The data evaluation module is used to retrieve each piece of data to be detected, perform semantic analysis on the data to be detected, generate one or more keywords, match the keywords with the sensitive word library, and output a yes or no matching result; The step of matching the keyword with a sensitive word library includes: Obtain the keyword, and obtain the context of the keyword; Retrieving the sensitive word library, comparing the sensitive word library with the keyword and the context of the keyword, and obtaining the number of matching words; Calculate the ratio of the number of matching words to the keyword and the context of the keyword; According to a preset threshold of the ratio, determining whether the keyword matches the sensitive word; A data analysis module, used to call a data analysis model, and when the matching result is negative, input the data to be detected into the data analysis model, analyze the sensitive features of the data to be detected, and output the analysis result; A data reporting module, used to generate an evaluation report based on the matching results and the analysis results; Wherein, the data analysis model includes: A data input unit, used to input all keywords of the data to be detected; An assignment unit, used for performing semantic assignment based on the keywords, respectively, wherein the semantic assignment refers to parsing the semantics of the keywords and labeling the semantics of the keywords, wherein the labeling is the assignment; A combination identification unit, based on the assignment, performs sensitive semantic combination to obtain an assignment field, and performs sensitive semantic identification on a plurality of randomly combined assignment fields; the sensitive semantic combination includes randomly extracting a plurality of the assignments and randomly combining the assignments; The sensitive semantic identification of the multiple randomly combined assigned fields includes a calculation formula for the score P of whether a certain assigned field is a sensitive feature, and the formula is as follows: ; Wherein, P is the score, ranging from 0 to 1, Q is the number of randomly combined field assignments, M is the number of matched sensitive features, and It is the total number of sensitive features of each sensitive information in the matching table; when the score P is greater than 0.3, it is considered that the data to be detected has sensitive features.

2. The data transaction compliance detection system according to claim 1, characterized in that: The sensitive word library is stored in a read-only memory.

3. The data transaction compliance detection system according to claim 1, characterized in that: Also includes: The authentication module is used to obtain the matching result and the analysis result, and generate an authentication mark according to the matching result and the analysis result.

4. The data transaction compliance detection system according to claim 1, characterized in that: Also includes: The information recommendation module is used to match the recommended information from the database according to the evaluation report and write the recommended information into the evaluation report.

5. A data transaction compliance detection method, characterized in that: include: Collect the data to be tested, including data directory information, transaction user information, data source information, and data authorization information; Retrieve each piece of data to be tested, perform semantic analysis on the data to be tested, generate one or more keywords, match the keywords with the sensitive word library, and output a yes or no match result; The step of matching the keyword with a sensitive word library includes: Obtain the keyword, and obtain the context of the keyword; Retrieving the sensitive word library, comparing the sensitive word library with the keyword and the context of the keyword, and obtaining the number of matching words; Calculate the ratio of the number of matching words to the keyword and the context of the keyword; According to a preset threshold of the ratio, determining whether the keyword matches the sensitive word; Retrieving a data analysis model, and when the matching result is negative, inputting the data to be detected into the data analysis model, analyzing the sensitive features of the data to be detected, and outputting the analysis result; Wherein, the data analysis model includes: A data input unit, used to input all keywords of the data to be detected; An assignment unit, used for performing semantic assignment based on the keywords, respectively, wherein the semantic assignment refers to parsing the semantics of the keywords and labeling the semantics of the keywords, wherein the labeling is the assignment; A combination identification unit, based on the assignment, performs sensitive semantic combination to obtain an assignment field, and performs sensitive semantic identification on a plurality of randomly combined assignment fields; the sensitive semantic combination includes randomly extracting a plurality of the assignments and randomly combining the assignments; The sensitive semantic identification of the multiple randomly combined assigned fields includes a calculation formula for the score P of whether a certain assigned field is a sensitive feature, and the formula is as follows: ; Wherein, P is the score, ranging from 0 to 1, Q is the number of randomly combined field assignments, M is the number of matched sensitive features, and is the total number of sensitive features of each sensitive information in the matching table; when the score P is greater than 0.3, it is considered that the data to be detected has sensitive features; An evaluation report is generated according to the matching result and the analysis result.

6. The data transaction compliance detection method according to claim 5 is characterized in that: The sensitive word library is stored in a read-only memory.

7. The data transaction compliance detection method according to claim 5 is characterized in that: Also includes: Acquire multiple data information, and cross-judge the multiple data information, including: respectively identifying identity attribute information of each data information to form an identity data set; The age range is determined according to the identity data set.

8. The data transaction compliance detection method according to claim 5 is characterized in that: Also includes: According to the evaluation report, recommendation information is matched from a database, and the recommendation information is written into the evaluation report.

Images