Intelligent data asset compliance risk analysis method, verification method based on intelligent data asset compliance risk analysis and computer program product

Abstract

The invention discloses an intelligent data asset compliance risk analysis method, a verification method based on intelligent data asset compliance risk analysis and a computer program product. The method comprises the following steps: determining a corresponding data source list from a database according to an instruction input by a user; performing data asset information extraction on the data source selected by the user, and determining missing data asset compliance features of the data source according to the extracted data asset information and preset data asset compliance feature requirements; identifying missing data asset compliance features by using a general large model and / or a preset compliance risk knowledge base according to the extracted data asset information; according to a preset compliance risk knowledge base, compliance information is supplemented for the data source selected by the user; according to a preset law compliance knowledge base and / or compliance risk knowledge base, compliance risk labels are supplemented for the data sources selected by the user; and writing the obtained data asset compliance risk analysis result into a data asset directory. According to the invention, data asset compliance information identification, compliance information supplementation and compliance risk label supplementation can be automatically carried out on the data source, intelligent data asset compliance risk analysis is realized, and the data asset compliance management efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the fields of big data, artificial intelligence and data asset technology, and in particular to an intelligent data asset compliance risk analysis method, a verification method based on intelligent data asset compliance risk analysis and a computer program product. Background Art

[0002] In the digital economy era, data has become an important part of corporate competitiveness. By collecting, analyzing and using data, companies can better understand market demand, optimize products and services, improve production efficiency and innovation, and thus gain advantages in the fierce market competition. Data assets can help companies identify market trends, predict consumer behavior, and then formulate more accurate marketing strategies and product planning, enhance market competitiveness, and bring value and benefits to companies. At the same time, data assets also face compliance management challenges such as security and privacy. Companies need to take effective measures to protect the security and integrity of data assets and avoid data leakage and loss. Data asset compliance risk refers to the risks faced by companies due to violations of relevant laws, regulations, industry standards and internal policies during the collection, storage, processing, transmission and use of data assets. These risks may lead to serious consequences such as legal penalties, reputation loss, and economic losses. Therefore, conducting data asset compliance risk analysis is crucial for the steady development and long-term competitiveness of companies. Data asset compliance risk analysis is traditionally performed manually, requiring manual data asset compliance business classification indexing and data cataloging to meet the needs of rapid verification of data asset compliance. Manual processing is time-consuming and labor-intensive, inefficient, and prone to errors, resulting in inaccurate compliance records of data assets. Summary of the invention

[0003] The purpose of the present invention is to provide an intelligent data asset compliance risk analysis method, a verification method based on intelligent data asset compliance risk analysis, and a computer program product that implements any of the above methods when executed. The method can realize intelligent data compliance risk analysis and improve the efficiency and accuracy of data asset compliance management.

[0004] In order to achieve the above object, the present invention provides an intelligent data asset compliance risk analysis method, comprising the following steps:

[0005] S101, accepting a user's dialogue interaction via text or voice, allowing the user to input instructions;

[0006] S102, according to the instruction input by the user, determining a corresponding data source list from the database for the user to select;

[0007] S103, performing compliance risk analysis on the data source selected by the user, including extracting data asset information from the data source selected by the user, and determining the missing data asset compliance features of the data source based on the extracted data asset information and preset data asset compliance feature requirements, where the data asset compliance feature requirements include asset name, asset category, asset table description, data source information, data content information, data processing information, data management information and data operation information; using a general big model and / or a preset compliance risk knowledge base to identify the missing data asset compliance features based on the extracted data asset information.

[0008] S104: According to the preset compliance risk knowledge base, the compliance information is supplemented for the data source selected by the user, and the compliance information includes laws, regulations and / or technical standards; specifically, the correlation between the instructions input by the user and each data source is calculated according to the VSM vector space model, and the data sources whose correlation reaches the preset threshold are listed from high to low according to the correlation for the user to choose.

[0009] S105: According to the preset legal compliance knowledge base and / or compliance risk knowledge base, a compliance risk label is added to the data source selected by the user. The compliance risk label includes data classification and grading, sensitivity level, data quality requirements, business process node, business entity and data responsible department.

[0010] S106, writing the obtained data asset compliance risk analysis results into the data asset directory, wherein the data asset compliance risk analysis results include the data source and its data asset information, compliance information and compliance risk labels.

[0011] S107, perform compliance check on the data of the data source in the data asset compliance risk analysis results according to preset data compliance rules, wherein the data compliance rules include data source compliance rules, data content compliance rules, data processing compliance rules, data management compliance rules and data operation compliance rules; and / or associate the compliance risk label in the data asset compliance risk analysis results of the data source selected by the user with the data asset directory where the data asset compliance risk analysis results are located.

[0012] S108, after outputting the data asset compliance risk analysis results to the user, output guidance information for the user to select additional compliance operations from, the guidance information includes additional data asset compliance feature requirements, additional compliance information items and / or additional compliance risk label items; the updated data asset compliance risk analysis results are obtained based on the user's selection.

[0013] The present invention also provides a verification method based on intelligent data asset compliance risk analysis, which performs compliance analysis on data according to the intelligent data asset compliance risk analysis method as described above, and then performs the following steps:

[0014] S201, accepting a user's dialogue interaction via text or voice, allowing the user to input a data asset verification instruction;

[0015] S202, performing data processing on the data asset verification instruction input by the user, verifying and obtaining corresponding data from the data asset catalog, and calling corresponding data analysis methods and / or data mining models in a pre-established data compliance model library to process the data, wherein the data compliance model library includes a variety of commonly used data analysis methods and a variety of commonly used data mining models;

[0016] S203, output the obtained data verification results to the user, and output the data asset association panorama to the user, wherein the asset association panorama includes the basic information of the table where the data is located, related attributes, table field information, table association relationship, and the blood relationship between the table and the field.

[0017] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements the intelligent data asset compliance risk analysis method and the verification method based on the intelligent data asset compliance risk analysis as described above.

[0018] The present invention provides an intelligent data asset compliance risk analysis method, a verification method based on intelligent data asset compliance risk analysis, and a computer program product, which can automatically identify data asset compliance information, supplement compliance information, and supplement compliance risk labels for data sources, thereby realizing intelligent data asset compliance risk analysis and improving the efficiency and accuracy of data asset compliance management. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flowchart of the intelligent data asset compliance risk analysis method.

[0020] Figure 2 It is a flowchart of the verification method based on intelligent data asset compliance risk analysis.

[0021] Figure 3 It is the structural block diagram of the intelligent data asset compliance risk analysis system. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0024] Intelligent data asset compliance risk analysis system Figure 3 As shown, it includes a user interface, a data asset compliance analysis platform and a data asset database. The data asset compliance analysis platform is connected to the user interface and the data asset database in communication. The data asset compliance analysis platform includes a processor and a computer-readable storage medium. The computer-readable storage medium stores a computer program product. The data asset compliance analysis platform processor executes the computer program product to achieve the following Figure 1 The intelligent data asset compliance risk analysis method shown or the implementation as follows Figure 2 The verification method for intelligent data asset compliance risk analysis is shown.

[0025] In this embodiment, the above-mentioned computer program product is deployed as an intelligent agent in the data asset compliance analysis platform. The data asset compliance analysis platform implements these two methods by running the intelligent agent. The following text illustrates the execution process of these two methods through a data asset compliance risk analysis example.

[0026] The process of intelligent data asset compliance risk analysis is as follows:

[0027] Users use the intelligent data asset compliance risk analysis system to conduct data asset compliance risk analysis and upload some data sources that the enterprise needs to analyze into the data asset database.

[0028] Preferably, technical personnel can pre-deploy data automation collection tools on the data asset compliance analysis platform, so that the data asset compliance analysis platform can automatically collect various types of data sources and store them in the data asset database as data sources to be analyzed.

[0029] To conduct data asset compliance risk analysis on these data sources to be analyzed, the user logs in to access the data asset compliance analysis platform through its user interface, and then inputs analysis instructions to the data asset compliance analysis platform through dialogue interaction on the user interface, such as voice input of "Please help me analyze the data related to the logistics domain". In other embodiments, the user can instead input "Please help me analyze the data related to the logistics domain" in text.

[0030] After the data asset compliance analysis platform receives the analysis instruction "Please help me analyze the logistics domain related data" input by the user, it determines the corresponding data source list from the data asset database for the user to choose. Specifically, the data asset compliance analysis platform calculates the correlation between the analysis instruction input by the user and the various data sources in the database according to the VSM vector space model, and then lists the data sources with a correlation reaching a preset threshold, such as 90%, from high to low according to the correlation for the user to choose. The similarity calculation process is: the metadata information of the data source and the data source keyword "logistics domain" in the analysis instruction are respectively represented as vectors, and the correlation between them is calculated using the following cosine similarity formula.

[0031] Cosine similarity formula:

[0032] If the data asset compliance analysis platform calculates that there is only one data source with a relevance of 90%, then the data source is directly selected and information is output to inform the user of the selected data source.

[0033] After the user selects the data source to be analyzed, the data asset compliance analysis platform automatically analyzes the data source selected by the user. The analysis process is as follows: The data asset compliance analysis platform extracts data asset information from the data source selected by the user. For example, the current data asset information of the data source extracted includes: asset name and data source information.

[0034] Data asset compliance risk analysis usually requires the following data asset information: asset name, asset category, asset table description, data source information, data content information, data processing information, data management information and data operation information. Therefore, this embodiment uses the above-mentioned data asset information as the data asset compliance feature requirements for data asset compliance risk analysis, that is, the data asset information type requirements include asset name, asset category, asset table description, data source information, data content information, data processing information, data management information and data operation information.

[0035] Non-preferably, in other embodiments, the data asset information type requirement can be changed to include only one or more of the asset name, asset category, asset table description, data source information, data content information, data processing information, data management information and data operation information.

[0036] After the data asset compliance analysis platform extracts data asset information from the data source, it determines the data asset compliance features that are missing from the data source based on the extracted data asset information and the preset data asset compliance feature requirements.

[0037] As mentioned above, the data asset information extracted by the data asset compliance analysis platform is the asset name and data source information. After comparing the extracted data asset information with the data asset compliance feature requirements, the data asset compliance analysis platform determines that the data source is missing the data asset information of the data asset compliance feature requirements, including asset category, asset table description, data content information, data processing information, data management information and data operation information.

[0038] Therefore, a general large model and / or a preset compliance risk knowledge base is used to identify missing data asset compliance features based on the extracted data asset information.

[0039] The identification process is as follows: First, the data asset compliance analysis platform uses the general big model on the market, such as Baidu's Wenxin Yiyan model, to supplement some data asset information that can be directly obtained based on the extracted data asset information according to general knowledge. For example, for semantic recognition, the data asset compliance analysis platform uses the general big model to identify the asset category through the asset name, thus obtaining the data asset information of the asset category; then, for some data asset information that cannot be directly obtained by the general big model according to general knowledge, the data asset compliance analysis platform uses the pre-built compliance risk knowledge base and the extracted data asset information to supplement the missing data asset compliance features. For example, according to the asset name, the asset table description corresponding to the asset name is searched from the compliance risk knowledge base, and according to the data source information, the data content information corresponding to the data source information is searched from the compliance risk knowledge base, and further according to the data content information, the data processing information, data management information and data operation information corresponding to the data content information are queried from the compliance risk knowledge base, thus obtaining five data asset information items: asset table description, data content information, data processing information, data management information and data operation information.

[0040] The above-mentioned compliance risk knowledge base is pre-constructed by legal experts. Legal experts collect relevant legal and regulatory documents of the industry and the enterprise itself, such as legal documents, standard specification documents, process documents, business glossary, etc., and then extract and abstract these industry business documents to summarize compliance knowledge, and build the enterprise compliance risk knowledge base based on this. Business knowledge includes the correspondence between asset names and asset table descriptions, the correspondence between data source information and data content information, and the correspondence between data content information and data processing information, data management information, and data operation information. Business knowledge also includes conventional business knowledge such as business field classification, business standards, business processes, and business entity categories, which are not exhaustive in this embodiment.

[0041] After the data asset compliance analysis platform identifies the data asset information for the data source to be analyzed (i.e., the data source selected by the user), it supplements the compliance information for the data source according to the preset compliance risk knowledge base. The compliance information includes business areas and business standards. The data asset compliance analysis platform matches the data asset information of the data source, such as asset name, asset category, or asset table description, with the classification rules of each business area in the compliance risk knowledge base to determine the business area to which the data source belongs, and further determines its business standards based on the business area to which the data source belongs. In other non-preferred embodiments, the compliance information can be changed to include only business areas or only business standards.

[0042] The data asset compliance analysis platform adds compliance risk labels to the data sources selected by users based on the preset legal compliance knowledge base and compliance risk knowledge base. The compliance risk labels include data classification and grading, sensitivity level, data quality requirements, business process nodes, business entities and data responsible departments.

[0043] The supplementary process is as follows: the data asset compliance analysis platform first determines the data classification and grading to which the above-mentioned data source belongs according to the data classification and grading standards in the legal compliance knowledge base; then determines the sensitivity level of each field in the above-mentioned data source according to the data management grading standards in the legal compliance knowledge base; then determines the data quality requirements of the above-mentioned data source according to the data quality standards in the legal compliance knowledge base; then determines the source department of the above-mentioned data source according to the data source department in the legal compliance knowledge base, and takes its source department as its data responsible department; then determines the business process node to which the above-mentioned data source belongs according to the business process in the compliance risk knowledge base; then determines the business entity to which the above-mentioned data source belongs according to the business entity category in the compliance risk knowledge base.

[0044] The legal compliance knowledge base is pre-constructed by legal experts and information technology personnel. Specifically, on the one hand, information technology personnel diagnose the compliance of user data assets from four aspects: current situation investigation and analysis, data asset management system planning, implementation path planning, and project implementation. On the other hand, they collect common legal compliance knowledge in the industry (such as national data security standards, data security laws, data classification and grading standards, data management grading standards) and corporate rules and regulations (such as XX company data security system, XX company data security training video, XX company data security standards, XX company data management grading standards, XX company data quality standards, XX company data source department), and then construct a legal compliance knowledge base based on the user data asset diagnosis and legal compliance knowledge.

[0045] Not preferably, in other embodiments, the compliance risk label can be changed to include only one or more of the data classification and grading, sensitivity level, data quality requirements, business process node, business entity and data responsible department.

[0046] According to the above process, the data asset compliance analysis platform analyzes and obtains the data asset information, compliance information and compliance risk labels of the data source, that is, the data asset compliance risk analysis results of the data source are obtained. The data asset compliance risk analysis results include the data source and its data asset information, compliance information and compliance risk labels.

[0047] The data asset compliance analysis platform outputs the data asset compliance risk analysis results to the user on the user interface through interactive dialogue. For example, the data asset compliance risk analysis result link is returned to the user online in the dialog box, so that the user can correct and confirm the data asset compliance risk analysis results and determine the accuracy of the data asset compliance risk analysis results. The user can click on the data asset compliance risk analysis result link to expand and view the data compliance risk analysis results to verify whether the data asset information, compliance information, and compliance risk labels of the data compliance risk analysis results are accurate.

[0048] Taking data asset information as an example, users browse and analyze the identified data asset information "asset category, asset table description, field Chinese, data processing information, data management information, data operation information". If the user believes that some data asset information is incomplete or incorrect, the user can directly change the data asset information in the expanded link. After the change is completed, save it, and the user interface will input the compliance risk analysis result change instruction to the data asset compliance analysis platform.

[0049] After the data asset compliance analysis platform receives the compliance risk analysis result change instruction input by the user, it updates the data asset compliance risk analysis result according to the change information in the compliance risk analysis result change instruction, and then writes the updated data asset compliance risk analysis result into the data asset directory of the data asset database.

[0050] If the user does not change the compliance risk analysis results, the data asset compliance analysis platform will write the data asset compliance risk analysis results automatically analyzed into the data asset directory of the data asset database.

[0051] Taking into account that the user may not immediately view the data asset compliance risk analysis result link returned by the data asset compliance analysis platform, in this embodiment, if the data asset compliance analysis platform does not receive the compliance risk analysis result change instruction input by the user for more than a preset time period, for example, more than 10 minutes, it will directly write the data asset compliance risk analysis results obtained by automatic analysis into the data asset directory of the data asset database.

[0052] If the user wants to view the data asset compliance risk analysis results later, he / she can use the dialogue interaction method on the user interface to enter "Please help me browse the data asset compliance risk analysis results of xxx data". The data asset compliance analysis platform will return the data asset compliance risk analysis result link to the user online in the dialog box. The user clicks the data asset compliance risk analysis result link to expand and view the data compliance risk analysis results.

[0053] For example, if the user believes that some data asset information is incomplete or incorrect, the user can directly change the data asset information in the expanded link, save the change after completion, and the user interface will input the compliance risk analysis result change instruction to the data asset compliance analysis platform.

[0054] After the data asset compliance analysis platform receives the compliance risk analysis result change instruction input by the user, it updates the data asset compliance risk analysis result according to the change information in the compliance risk analysis result change instruction, and then updates the original data asset compliance risk analysis results in the data asset catalog according to the updated data asset compliance risk analysis result.

[0055] The data asset compliance analysis platform analyzes all data sources according to unified analysis standards. Since users may want to conduct a more comprehensive data asset compliance risk analysis on certain special data sources, when browsing the data asset compliance risk analysis results returned by the data asset compliance analysis platform, they may think that the data asset compliance risk analysis results are not comprehensive / complete enough and need to change the data asset compliance risk analysis results.

[0056] Taking into account that it is time-consuming and laborious for users to manually change the data asset compliance risk analysis results as described above, in this embodiment, preferably, the data asset compliance analysis platform outputs secondary analysis guidance information while outputting the data asset compliance risk analysis results to the user, that is, attaching the secondary analysis guidance information in the data asset compliance risk analysis result link.

[0057] The secondary analysis guidance information is intended to provide more secondary analysis items for users to choose from based on the original analysis standards of data asset information, compliance information, and compliance risk labels. The guidance information includes additional data asset compliance feature requirements, additional compliance information items, and / or additional compliance risk label items.

[0058] In this way, users can select the data asset compliance features, compliance information and / or compliance risk labels that need to be added and completed from the secondary analysis guidance information. If the data asset compliance analysis platform receives the secondary analysis additions input by the user (i.e., the data asset compliance feature requirements, compliance information and / or compliance risk label additions), it will re-analyze according to the above automatic analysis process based on the secondary analysis additions to obtain updated data asset compliance risk analysis results.

[0059] For example, if the user inputs additional data asset compliance feature requirements and additional compliance risk label items, the data asset compliance analysis platform will add the additional data asset compliance feature requirements input by the user on the basis of the original data asset compliance feature requirements, and will add the additional compliance risk label items input by the user on the basis of the original compliance risk label categories. In this way, in the re-analyzed data asset compliance risk analysis results, the data asset information will include the original data asset compliance features and the above-mentioned additional data asset compliance feature requirements, and the compliance risk labels will include the original compliance risk labels and the above-mentioned additional compliance risk labels.

[0060] Preferably, in order to ensure the accuracy of the data compliance risk analysis results, the data asset compliance analysis platform performs a compliance check on the data of the data source in the data asset compliance risk analysis results according to preset data compliance rules before writing the data compliance risk analysis results into the data asset directory. If it is non-compliant, a prompt will be given for the non-compliant data.

[0061] Compliance inspections mainly include the following five aspects:

[0062] 1. Data source compliance check: Check the data type, length, data format and other aspects based on the analyzed data asset information to ensure compliance with data specifications, and check whether the data acquisition behavior violates any laws and regulations, national policies and social public morality, and does not infringe on the legitimate rights of any third party;

[0063] 2. Data content compliance check: Check the data content to ensure that the data value is reasonable and correct. Check that the data content is true, legal, and compliant. Do not store illegal data that is not allowed to be collected or stored by laws and regulations.

[0064] 3. Data processing compliance check: Check whether there is any data missing, ensure that the data processing behavior does not violate relevant legal provisions and complies with the principles of legality, legitimacy and necessity;

[0065] 4. Data management compliance inspection: Check whether the data is included in data management to ensure data management security. Establish data compliance management systems in accordance with laws, regulations and rules. Carry out management activities including compliance management system establishment, risk identification, risk assessment and disposal. Establish corresponding full-chain supervision and management mechanisms in the fields of data classification and grading management, cross-border data, and personal information protection.

[0066] 5. Data operation compliance inspection: Carry out data operation business in accordance with the law, obtain corresponding qualifications, administrative licenses and full authorization, establish a sound internal control system, and ensure that data operation business does not endanger national security, public interests, or infringe upon the legitimate rights and interests of individuals and organizations.

[0067] This embodiment pre-constructs corresponding compliance rules according to the compliance checks in the five aspects, specifically: regarding data source compliance check, construct data source compliance rules (including multiple compliance rules); regarding data content compliance check, construct data content compliance rules (including multiple compliance rules); regarding data processing compliance check, construct data processing compliance rules (including multiple compliance rules); regarding data management compliance check, construct data management compliance rules (including multiple compliance rules); regarding data operation compliance check, construct data operation compliance rules (including multiple compliance rules). The above five compliance rules are used as preset data compliance rules.

[0068] If the data asset compliance analysis platform detects that the data asset compliance risk analysis results do not comply with a certain compliance rule, it will output the compliance rule and the data that does not comply with the compliance rule to the user, prompting the user that the data is non-compliant.

[0069] Users can manually process non-compliant data to make it meet compliance requirements, and then upload it to the data asset compliance analysis platform.

[0070] The data asset compliance analysis platform will re-check the compliance of the data processed by the user, and if it complies with the rules, it will be updated to the data asset compliance risk analysis results.

[0071] After all the data in the data source of the data asset compliance risk analysis results are compliant, the data asset compliance analysis platform writes the data asset compliance risk analysis results into the data asset directory of the database.

[0072] Specifically, the data asset directory includes multiple subject directories. The data asset compliance analysis platform first determines the subject directory to which the data asset compliance risk analysis results belong, and then writes the data asset compliance risk analysis results into the corresponding subject directory.

[0073] After the data asset compliance analysis platform writes the data asset compliance risk analysis results into the data asset directory, it associates the compliance risk tags in the data asset compliance risk analysis results with the subject directory of the data asset directory where the data asset compliance risk analysis results are located. In this way, the data asset directory and the compliance risk tags form a network data retrieval system.

[0074] In this embodiment, the data asset compliance analysis platform can also use the legal compliance knowledge base to provide users with data asset compliance risk analysis consulting services. If the user wants to consult about data asset compliance risk analysis, he / she can input the question to be consulted through dialogue interaction on the user interface, such as "how to analyze". After the data asset compliance analysis platform receives the analysis consulting information "how to analyze", it obtains the corresponding consulting answer from the legal compliance knowledge base based on the keywords "analysis" and "how to do" in the analysis consulting information and outputs it to the user. For example, it extracts the implementation process of the analysis from the legal compliance cases or legal compliance implementation charter documents in the legal compliance knowledge base, and outputs it to the user as the consulting answer.

[0075] As mentioned above, the data asset compliance analysis platform has realized intelligent data asset compliance risk analysis. On this basis, it can also realize the verification and use of data assets based on intelligent data asset compliance risk analysis. The process is as follows:

[0076] If a user wants to view a certain data source, he / she logs in to access the data asset compliance analysis platform through its user interface, and then enters data asset verification instructions to the data asset compliance analysis platform through dialogue interaction on the user interface, such as voice input of "Please help me verify the relevant asset table of inventory items."

[0077] The data asset compliance analysis platform receives the data asset verification instruction "Please help verify the relevant asset table of inventory goods" input by the user. Since the data asset verification instruction "Please help verify the relevant asset table of inventory goods" input by the user does not need to analyze and process the data, the result data required by the user can be obtained directly from the data asset catalog. Therefore, the data asset compliance analysis platform verifies the corresponding data from the data asset catalog based on the asset keywords "inventory" and "goods" in the verification instruction, and outputs the data and its asset association panorama to the user. The asset association panorama includes the basic information of the table where the data is located, related attributes, table field information, table association relationship, and the blood relationship between the table and the field.

[0078] If the data asset verification instruction input by the user requires analysis and processing of the data, the data asset compliance analysis platform verifies the corresponding data from the data asset directory, calls the corresponding data analysis method and / or data mining model in the pre-established data compliance model library to calculate the data, obtains the data calculation results and outputs the data calculation results to the user.

[0079] The data compliance model library includes a variety of commonly used data analysis methods such as statistical analysis, reliability analysis, correlation analysis, regression analysis, discriminant analysis, factor analysis, as well as a variety of commonly used data mining models such as linear regression, logistic regression, decision tree, naive Bayes, k-means, etc.

[0080] The data compliance model library is pre-built by legal experts and technicians, and each data analysis method and each data mining model is encapsulated into an API tool for the data asset compliance analysis platform to directly call and process the data.

[0081] For example, if a user enters a data asset verification instruction "Help me verify the inventory of products with annual sales exceeding 10,000 pieces and unit prices exceeding 50,000 yuan", the data asset verification instruction requires compliance analysis and processing of the data. Therefore, the data asset compliance analysis platform first verifies the corresponding data from the data asset catalog based on the asset keywords "sales", "unit price", "inventory", and "product" in the verification instruction. For example, it verifies the product sales statistics table (the table includes fields such as sales, unit price, inventory status, and product name), and then calls the decision tree model from the data analysis and mining model library to filter the product sales statistics table, obtains the data processing results, and outputs the data processing results to the user.

[0082] If the user wants to generate a visualization chart based on the data returned by the data asset compliance analysis platform, he can continue to enter the visualization chart generation instruction through dialogue interaction on the user interface, such as entering "please generate a pie chart based on the above data". After receiving the visualization chart generation instruction from the user, the data asset compliance analysis platform generates a corresponding visualization chart based on the output data processing results or data, that is, a pie chart, and outputs it to the user.

[0083] The specific process by which the data asset compliance analysis platform verifies the corresponding data from the data asset catalog based on the data asset verification instructions input by the user is: first, a corresponding SQL statement is generated based on the data asset verification instructions input by the user, and then the SQL statement is sent to the data asset database for execution, thereby verifying the corresponding data from the data asset catalog.

[0084] Among them, the SQL statement generation process is as follows: the data asset compliance analysis platform first determines the corresponding data source in the data asset directory based on the data asset verification instruction keywords entered by the user, and then generates the corresponding SQL statement based on the data asset verification instruction entered by the user and the data asset information of the data source.

[0085] If the data asset verification instruction entered by the user contains business indicators, for example, the user enters the data asset verification instruction "this month's product shipments", "this month's product shipments" is a business indicator. Therefore, after determining the data source and generating the corresponding SQL statement, the data asset compliance analysis platform further combines the metadata information of the data source and the business indicator definition information (field name, qualifier, aggregation function, dimension, etc.) to further improve the generated SQL statement and obtain the improved SQL statement.

[0086] In this embodiment, the data asset compliance analysis platform uses the data asset verification instruction and the final SQL statement input by the user as a historical question and answer record, and builds a historical question and answer corpus based on this. After receiving the data asset verification instruction input by the user, the data asset compliance analysis platform first queries the historical question and answer prediction library to see whether there is a historical question and answer record with the same data asset verification instruction. If so, it directly obtains the SQL statement of the historical question and answer record, sends it to the data asset database for execution, and returns it to the user after obtaining the corresponding data. In this way, the data asset compliance analysis platform does not need to generate the corresponding SQL statement again, and can quickly verify the data required by the user, thereby improving the efficiency of data asset verification.

[0087] The above is only an implementation method of the invention, and does not limit the scope of patent protection. Those skilled in the art can make non-substantial changes or substitutions based on the invention, which still fall within the scope of patent protection.

Claims

1. An intelligent data asset compliance risk analysis method, characterized by: The steps include: S101, accepting a user's dialogue interaction via text or voice, allowing the user to input instructions; S102, according to the instruction input by the user, determining a corresponding data source list from the database for the user to select; S103, performing compliance risk analysis on the data source selected by the user, including extracting data asset information from the data source selected by the user, and determining the missing data asset compliance features of the data source based on the extracted data asset information and preset data asset compliance feature requirements, where the data asset compliance feature requirements include asset name, asset category, asset table description, data source information, data content information, data processing information, data management information and data operation information; using a general big model and / or a preset compliance risk knowledge base to identify the missing data asset compliance features based on the extracted data asset information.

2. S104, based on the preset compliance risk knowledge base, supplement the compliance information for the data source selected by the user, the compliance information includes laws, regulations and / or technical standards; specifically, calculate the correlation between the instructions input by the user and each data source based on the VSM vector space model, and list the data sources whose correlation reaches the preset threshold from high to low according to the correlation for the user to choose.

3. S105, based on the preset legal compliance knowledge base and / or compliance risk knowledge base, add compliance risk labels for the data source selected by the user. The compliance risk labels include data classification and grading, sensitivity level, data quality requirements, business process nodes, business entities and data responsible departments; S106, writing the obtained data asset compliance risk analysis results into the data asset directory, where: The data asset compliance risk analysis results include data sources and their data asset information, compliance information, and compliance risk labels.

4. S107, perform compliance check on the data of the data source in the data asset compliance risk analysis results according to preset data compliance rules, where the data compliance rules include data source compliance rules, data content compliance rules, data processing compliance rules, data management compliance rules and data operation compliance rules; and / or associate the compliance risk label in the data asset compliance risk analysis results of the data source selected by the user with the data asset directory where the data asset compliance risk analysis results are located. 5.S108. After the data asset compliance risk analysis results are output to the user, guidance information is output for the user to select additional compliance operations from. The guidance information includes additional data asset compliance feature requirements, additional compliance information and / or additional compliance risk labels. The updated data asset compliance risk analysis results are obtained based on the user's selection.

6. A verification method based on intelligent data asset compliance risk analysis, characterized in that: Perform compliance analysis on the data according to the intelligent data asset compliance risk analysis method as required by Right 1, and then perform the following steps: S201, accepting a user's dialogue interaction via text or voice, allowing the user to input a data asset verification instruction; S202, performing data processing on the data asset verification instruction input by the user, verifying and obtaining corresponding data from the data asset catalog, and calling corresponding data analysis methods and / or data mining models in a pre-established data compliance model library to process the data, wherein the data compliance model library includes a variety of commonly used data analysis methods and a variety of commonly used data mining models; S203, output the obtained data verification results to the user, and output the data asset association panorama to the user, wherein the asset association panorama includes the basic information of the table where the data is located, related attributes, table field information, table association relationship, and the blood relationship between the table and the field.

7. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, it implements the intelligent data asset compliance risk analysis method as described in claim 1 or the verification method based on intelligent data asset compliance risk analysis as described in claim 2.

Images