Market supervision law enforcement data quality evaluation method and device, and storage medium
Through the market supervision and law enforcement data quality evaluation method, processors and storage media are used to solve the problem of silos of grassroots market supervision data, and quantitative evaluation of market supervision data is realized, providing a foundation for the construction of a digital government.
Patent Information
- Application Number
- CN202410842631.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-27
- Publication Date
- 2024-11-12
AI Technical Summary
The lack of unified data standards in grassroots market supervision, evidence collection and law enforcement systems has led to serious data silos and affecting the development of digital governments. A method of market supervision and law enforcement data quality evaluation is needed to achieve unified and effective data evaluation.
A method for evaluating data quality in market supervision law enforcement is proposed, including obtaining the data objects to be evaluated, selecting evaluation indicators, setting constraint rules, quantifying evaluation indicators and conducting quantitative evaluation indicators, adopting a complete evaluation index system and systematic evaluation plan, using processors and Storage media achieves data quality evaluation.
A complete evaluation index system and systematic evaluation plan have been built, and a quantitative assessment of the quality of market supervision data has been realized, providing a foundation for the construction of a digital government.
Smart Images

Figure CN118941312A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data quality evaluation, and in particular to a method, device and storage medium for evaluating market supervision and law enforcement data quality. Background Art
[0002] The grassroots market supervision, evidence collection and law enforcement system platforms lack unified data standards and design standards, and the problem of data silos between systems seriously affects the long-term development of digital government. Therefore, a unified and effective evaluation of market supervision and law enforcement data is a necessary path to establish a digital government. Summary of the invention
[0003] The purpose of the present invention is to propose a market supervision and law enforcement data quality evaluation method, device and storage medium for unified and effective evaluation of market supervision and law enforcement data.
[0004] The present invention provides a method for evaluating the quality of market supervision and law enforcement data, which specifically includes the following steps:
[0005] S1. Obtain the data object to be evaluated;
[0006] S2, selecting an evaluation indicator from the evaluation indicator set;
[0007] S3. According to the selected evaluation indicators, set constraint rules and quantify the evaluation indicators;
[0008] S4. Conduct quantitative evaluation of the data object based on the quantified evaluation indicators to form an evaluation result.
[0009] A storage medium, wherein the storage medium stores instructions and data for implementing a method for evaluating the quality of market supervision and law enforcement data.
[0010] A market supervision and law enforcement data quality evaluation device includes: a processor and a storage medium; the processor loads and executes instructions and data in the storage medium to implement a market supervision and law enforcement data quality evaluation method.
[0011] The beneficial effects provided by the present invention are: constructing a complete evaluation index system and a systematic evaluation plan, realizing the quantitative evaluation of the quality of market supervision data, and making certain basic construction for the subsequent unification of data and the construction of a digital government. BRIEF DESCRIPTION OF THE DRAWINGS
[0012] Figure 1 It is a schematic flow chart of the method of the present invention;
[0013] Figure 2 It is a working schematic diagram of the hardware device of the present invention. DETAILED DESCRIPTION
[0014] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described below with reference to the accompanying drawings.
[0015] Please refer to Figure 1 , Figure 1 : is a flow chart of the method of the present invention; the present invention provides a method for evaluating the quality of market supervision and law enforcement data, and the method specifically includes the following steps:
[0016] S1. Obtain the data object to be evaluated;
[0017] S2, selecting an evaluation indicator from the evaluation indicator set;
[0018] S3. According to the selected evaluation indicators, set constraint rules and quantify the evaluation indicators;
[0019] It should be noted that the evaluation indicators in step S2 include: integrity indicator, consistency indicator, accuracy indicator, machine readability indicator, timeliness indicator, accessibility indicator and security indicator.
[0020] It should be noted that in step S3, the constraint rules include: empirical analysis method, missing value analysis method, logical relationship analysis method, repeated value analysis method, comparison analysis method, experimental observation method, phrase comparison analysis method and range analysis method.
[0021] The empirical analysis method refers to comparing and verifying the data values with the empirical values generated in daily production and life to confirm or falsify the data content. The empirical analysis method is often used in conjunction with the logical relationship analysis method and the phrase comparison analysis method.
[0022] Missing value analysis refers to checking all data items of each data record one by one to identify whether the data items of the data record are empty.
[0023] Logical relationship analysis refers to analyzing the internal logical relationship between the values of related data items in data records to identify data records that violate logical relationships.
[0024] Duplicate value analysis refers to comparing the values of some or all data items in each data record with all data records one by one to identify duplicate records in a non-redundant manner.
[0025] Comparative analysis refers to directly comparing whether the data values are the same.
[0026] The experimental observation method refers to setting certain conditions, controlling the behavior process, and observing whether the results that meet the conditions appear.
[0027] The phrase comparison analysis method refers to comparing the value of the target data item in the data record with the target sensitive words, common wrong words and other vocabulary contents to identify data records that are the same as the vocabulary.
[0028] Range analysis refers to comparing the value of a data item with the value range of a field to identify whether it exceeds the value range.
[0029] In the present invention, different constraint rules and quantification standards are adopted according to different selected indicators.
[0030] As an embodiment, when the integrity index is taken as the primary index, it also has various secondary indicators under it, including: data set integrity index, identification subset metadata integrity index, content subset metadata integrity index, identification data record data item integrity index, content data record integrity index and content data record data item integrity index.
[0031] The data set integrity index refers to whether a data set is fragmented or artificially split into multiple data sets. The example / description of this index is as follows: the same data set is split into multiple data sets according to attributes such as time, region, type, etc. The evaluation method of the present invention is: empirical analysis method; specifically, the calculation method or quantification method is: when the data set is incomplete, the index item is 0, and when it is complete, the index item is 1.
[0032] The completeness index of the identifier subset metadata refers to whether the identifier subset metadata is complete. Examples / explanations of this index are as follows: a) the information summary is empty; b) the information subject is empty. The evaluation method of the present invention is: missing value analysis method; specifically, its calculation method or quantification method is: if the identifier subset metadata is incomplete, this item is 0; if the identifier subset metadata is complete, this item is 1.
[0033] Among them, the content subset metadata completeness index refers to whether the content subset metadata is complete; examples / explanations of this index are as follows: a) The name and address of the party are the metadata of the content subset of the "party information" data set; if the metadata of "party information" is missing, the use value of the data set is significantly reduced; b) A content subset with only one metadata is considered incomplete metadata; specifically, its calculation method or quantification method is: if the content subset metadata is incomplete, this item is 0. If the content subset metadata is complete, this item is 1.
[0034] Among them, the data item integrity index of the identification data record means that all data items of the identification data record are completed and not missing; the example / description of this index is as follows: in the "Party Information" data set, the case information summary content of the identification data record is empty. The evaluation method of the present invention is: missing value analysis method. Specifically, its calculation method or quantification method is: if the data item is missing, the item is 0, and if the data item is complete, the item is 1.
[0035] The content data record integrity index means that a data set covers all data records generated within the effective date. The example / explanation of this index is as follows: in the "Basic Information" data set, if the date is missing or is an unreasonable period, the data record is considered incomplete. Specifically, its calculation method or quantification method is: if the data record within the effective date is missing, this item is 0; if the data record within the effective date is complete, this item is 1.
[0036] The data item integrity index of the content data record refers to that the data items of each data record are complete and not missing. The example / explanation of this index is as follows: In the "Party Information" data set, a data record includes the basic data items of the party's name and address; all data records with missing party names or missing party addresses are incomplete; specifically, its calculation method or quantification method is: Where A = the number of data records with complete data items filled in, and B = the total number of data records.
[0037] Please refer to Table 1, which is the data integrity evaluation index.
[0038] Table 1 Integrity evaluation indicators
[0039]
[0040] As an embodiment, when the consistency index is taken as the first-level index, it has various second-level indicators under it, including: identification data record data item and metadata consistency index, content data record data item and metadata consistency index, content data record data item and metadata measurement unit consistency index and content data record consistency index.
[0041] Among them, the consistency index of the identification data record data item and metadata refers to the consistency of the identification data record data item and metadata; the example / explanation of this index is as follows: the update period in the identification subset metadata is filled in with the name of the case handling agency in the actual content. The evaluation method of the present invention is: logical relationship analysis method and empirical analysis method. Specifically, its calculation method or quantification method is: if the identification data record data item is inconsistent with the metadata, the item is 0, and if the identification data record data item is consistent with the metadata, the item is 1.
[0042] The consistency index of content data record data item and metadata refers to whether the content data record data item is consistent with the metadata. The example / description of this index is as follows: the content subset metadata is "name", which is filled in as "×× Co., Ltd." in the actual content; the evaluation method of the present invention is: logical relationship analysis method and empirical analysis method. Specifically, its calculation method or quantification method is: A = the number of data records with consistent data items in the content data record and metadata, B = the total number of data records;
[0043] Among them, the consistency index of content data record data item and metadata measurement unit means that the content data record data item is consistent with the metadata measurement unit. The example / explanation of this index is as follows: the measurement unit of content subset metadata is "10,000 yuan / year", and "yuan / year" is used to fill in the actual content. The evaluation method of the present invention is: logical relationship analysis method and empirical analysis method. Specifically, its calculation method or quantification method is: A = the number of data records whose content data record items are consistent with the metadata measurement unit, B = the total number of data records.
[0044] The content data record consistency index refers to the consistency and non-conflict of the content data records opened for the same matter within the same time period or fixed update frequency. The example / explanation of this index is as follows: In the "basic information", if there are two records of the same time and place but different contents, the data records are considered inconsistent; the evaluation methods of the present invention are: logical relationship analysis method and empirical analysis method. Specifically, its calculation method or quantification method is: A = the number of data records with consistent content data records, B = the total number of data records.
[0045] Please refer to Table 2, which is a schematic diagram of consistency evaluation indicators.
[0046] Table 2 Consistency evaluation indicators
[0047]
[0048] As an embodiment, when the accuracy index is taken as the primary index, there are various secondary indexes under it, including: data content correctness index, data format correctness index, and data uniqueness index.
[0049] The data content correctness index refers to whether the data content is the expected data. The evaluation method of the present invention is: comparative analysis method. Specifically, its calculation method or quantification method is: A = the number of elements in the data set that meet the data correctness requirements, B = the number of elements in the data set being evaluated.
[0050] The correctness index of the data format refers to whether the data format (including data type, value range, data length, etc.) meets the expected requirements. The example / description of this index is as follows: For example, the unified social credit code of the market entity does not meet the compilation requirements. The evaluation method of the present invention is: comparative analysis method. Specifically, its calculation method or quantification method is: A = the number of elements in the data set that meet the data format requirements, B = the number of elements in the data set being evaluated.
[0051] The data uniqueness index refers to the uniqueness and non-repetition of metadata of a data set or content subset. The example / description of this index is as follows: There are two data sets named "Party Information" with the same content, and the data sets are repeated. The evaluation method of the present invention is: repeated value analysis method. Specifically, the calculation method or quantification method is: if the metadata of a data set or content subset is repeated, this item is 0; if the metadata of a data set or content subset is not repeated, this item is 1.
[0052] Please refer to Table 3, which is a schematic diagram of accuracy evaluation indicators.
[0053] Table 3 Accuracy evaluation indicators
[0054]
[0055] As an embodiment, when the machine readability evaluation index is used as the first-level index, it also has a second-level index, which is the machine readability index of the data set. The machine readability index of the data set refers to that the downloadable data set format is easy to be read by the machine, such as CSV, JSON, XML, XLS, etc. Examples / explanations of this index are as follows: a) Providing data sets in non-machine-readable formats such as PDF and pictures; b) Converting data in unstructured DOC, PDF and other files into machine-readable formats such as XLS and CSV in a rigid way, while the data is still unstructured in essence; c) The provided data packet cannot be opened normally; d) There is no data in the provided data packet that can be called and the data set call fails; The evaluation method of the present invention is: experimental observation method. Specifically, the calculation method or quantification method is: if the data set is not machine-readable, the item is 0, and if the data set is machine-readable, the item is 1.
[0056] Please refer to Table 4, which is a diagram of machine readability evaluation indicators.
[0057] Table 4 Machine readability evaluation index
[0058]
[0059] As an embodiment, when the standardization evaluation index is taken as the first-level index, it also has various second-level indicators under it, including: content subset metadata standardization indicators, identification data record information resource name standardization indicators, identification data record data item format filling standardization indicators, content data record data item value standardization indicators and data record data item format filling standardization indicators.
[0060] Among them, the content subset metadata standardization index refers to the content subset metadata name being clear and unambiguous. The example / explanation of this index is as follows: the content subset metadata name contains words whose meaning cannot be directly understood, such as the metadata content is a name, and the metadata name is "xm"; the evaluation method of the present invention is: phrase comparison analysis method, empirical analysis method. Specifically, its calculation method or quantification method is: if the content subset metadata name is unclear and ambiguous, this item is 0; if the content subset metadata name is clear and unambiguous, this item is 1.
[0061] Among them, the indicator of the standardization of the name of the information resource of the data record refers to that the name of the information resource is clear and unambiguous. The example / explanation of this indicator is as follows: the content of the data set cannot be directly understood by the name, such as the name of the data set is "Public Information of Spot Check Result", and it is impossible to identify which field the spot check result information is for; the evaluation method of the present invention is: empirical analysis method. Specifically, the calculation method or quantification method is: if the name of the information resource is unclear and ambiguous, this item is 0; if the name of the information resource is clear and unambiguous, this item is 1.
[0062] Among them, the indicator of the standardization of the format of the data item of the identification data record means that the format of the data item of the identification data record is consistent with the metadata. The example / description of this indicator is as follows: the format of the release date of the identification subset metadata is required to be YYYY-MM-DD, and the actual format is MM-DD-YYYY; the evaluation method of the present invention is: empirical analysis method. Specifically, the calculation method or quantification method is: if the format of the identification subset metadata is inconsistent with the actual business, this item is 0; if the format of the identification subset metadata is consistent with the actual business, this item is 1.
[0063] Among them, the content data record data item value standardization index means that the content filled in the data item in the content data record is clear, has a clear dictionary meaning, and conforms to the actual business value range. Examples / explanations of this index are as follows: a) The data item "gender" has only "0, 1" values, and there is no corresponding dictionary table; b) The data item "age" has a value that exceeds the reasonable range; the evaluation method of the present invention is: value range analysis method. Specifically, its calculation method or quantification method is: A = the number of data records whose content data items are clear and in line with the actual value range, B = the total number of data records.
[0064] Among them, the indicator of the standardization of the name of the information resource of the data record refers to that the name of the information resource is clear and unambiguous. The example / explanation of this indicator is as follows: the content of the data set cannot be directly understood by the name, such as the name of the data set is "Public Information of Spot Check Result", and it is impossible to identify which field the spot check result information is for; the evaluation method of the present invention is: empirical analysis method. Specifically, the calculation method or quantification method is: if the name of the information resource is unclear and ambiguous, this item is 0; if the name of the information resource is clear and unambiguous, this item is 1.
[0065] Among them, the indicator of the standardization of the format of the data item of the identification data record means that the format of the data item of the identification data record is consistent with the metadata. The example / description of this indicator is as follows: the format of the release date of the identification subset metadata is required to be YYYY-MM-DD, and the actual format is MM-DD-YYYY; the evaluation method of the present invention is: empirical analysis method. Specifically, the calculation method or quantification method is: if the format of the identification subset metadata is inconsistent with the actual business, this item is 0; if the format of the identification subset metadata is consistent with the actual business, this item is 1.
[0066] Among them, the standardization indicator of filling in the format of recorded data items means that the format of the data items recorded by the content data is consistent with the metadata. The example / description of this indicator is as follows: the format of the release date of the content subset metadata is required to be YYYY-MM-DD, and the actual format is MM-DD-YYYY; the evaluation method of the present invention is: empirical analysis method. Specifically, its calculation method or quantification method is: A = the number of data records whose metadata format of the content subset is consistent with the actual business, B = the total number of data records.
[0067] Please refer to Table 5, which is a diagram of normative evaluation indicators.
[0068] Table 5 Normative evaluation indicators
[0069]
[0070]
[0071] As an embodiment, when the timeliness evaluation index is used as a primary index, it also has a secondary index, which is the data set timeliness index. The data set timeliness index means that a data set with timeliness should be uploaded or updated in a timely manner, and the uploaded data set is within the validity period; examples / explanations of this index are as follows: a) The uploaded data set data is expired; b) There is only an information directory, but no corresponding data set; the evaluation method of the present invention is: empirical analysis method. Specifically, the calculation method or quantification method is: if the data set is not uploaded in time within the validity period, this item is 0; if the data set is uploaded in time within the validity period, this item is 1.
[0072] Please refer to Table 6, which is a schematic diagram of timeliness evaluation indicators.
[0073] Table 6 Timeliness evaluation index
[0074]
[0075] As an example, when the accessibility evaluation index is used as a primary index, there are various secondary indexes under it, including: an accessibility index and a usability index.
[0076] Among them, the accessibility index refers to the availability of data when needed; the evaluation method of the present invention is: comparative analysis method. Specifically, its calculation method or quantification method is:
[0077] A = the number of elements in the dataset that meet the accessibility requirements, B = the number of elements in the dataset that are evaluated.
[0078] Among them, the usability index refers to the usability of the data within the set effective period; the evaluation method of the present invention is: comparative analysis method. Specifically, its calculation method or quantification method is: A = the number of elements in the dataset that meet the usability requirements, B = the number of elements in the dataset being evaluated.
[0079] Please refer to Table 7, which is a schematic diagram of accessibility evaluation indicators.
[0080] Table 7 Accessibility evaluation indicators
[0081]
[0082] As an embodiment, when the security evaluation index is marked as a first-level index, it also has a second-level index, which is a sensitive data item desensitization index. The sensitive data item desensitization index means that the disclosed data has been desensitized for sensitive data items and will not cause serious impact on public interests or personal interests; the evaluation method of the present invention is: phrase comparison method. Specifically, the calculation method or quantification method is: if the sensitive data item is not desensitized, the item is 0; if the sensitive data item is desensitized, the item is 1.
[0083] S4. Conduct quantitative evaluation of the data object based on the quantified evaluation indicators to form an evaluation result;
[0084] It should be noted that in step S4, the quantitative evaluation method includes: subjective evaluation method, subjective and objective combination method and objective evaluation method. Among them, the subjective evaluation method includes: expert meeting method, tomography analysis method, Delphi method. The subjective and objective combination method includes: Electre method, PROMETHEE method and fuzzy evaluation method. The objective evaluation method includes: entropy weight method, coefficient of variation method and negative correlation coefficient method.
[0085] It should be noted that in step S4, the quantitative evaluation of the data object includes: data set metadata evaluation, identification data record evaluation and content data record evaluation.
[0086] Specifically, dataset metadata evaluation refers to the analysis and evaluation of data quality based on indicators such as the completeness of the dataset identifier subset metadata, and the completeness, uniqueness, and standardization of the content subset metadata.
[0087] Specifically, identification data record evaluation refers to the data quality analysis and evaluation based on indicators such as the standardization of the identification data record information of the data set, as well as the integrity of the data items in the identification data record, the consistency between the data item content and metadata, and the standardization of the data item format.
[0088] Specifically, content data record evaluation refers to the data quality analysis and evaluation based on indicators such as the completeness, consistency, and accuracy of the content data records of the data set, as well as the completeness of the data items in the content data records, consistency with metadata, consistency with metadata measurement units, filling standardization, value standardization, and format filling standardization.
[0089] The results of the preliminary evaluation and detailed evaluation steps are summarized to form the evaluation results. If it is necessary to calculate the score, it is recommended to select appropriate evaluation indicators and set the score and weight according to actual needs, and score according to the percentage system. A score of 60 or more is considered qualified, and a score below 60 is considered unqualified.
[0090] Please refer to Figure 2 , Figure 2 It is a working diagram of the hardware device of an embodiment of the present invention, and the hardware device specifically includes: a market supervision and law enforcement data quality evaluation device 401, a processor 402 and a storage medium 403.
[0091] A market supervision and law enforcement data quality evaluation device 401: The market supervision and law enforcement data quality evaluation device 401 implements the market supervision and law enforcement data quality evaluation method.
[0092] Processor 402: The processor 402 loads and executes the instructions and data in the storage medium 403 to implement the market supervision and law enforcement data quality evaluation method.
[0093] Storage medium 403: The storage medium 403 stores instructions and data; the storage medium 403 is used to implement the market supervision and law enforcement data quality evaluation method.
[0094] In summary, the beneficial effects of the present invention are: a complete evaluation index system and a systematic evaluation scheme are constructed, quantitative evaluation of the quality of market supervision data is achieved, and a certain infrastructure is laid for the subsequent unification of data and the construction of a digital government.
[0095] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for evaluating the quality of market supervision and law enforcement data, characterized by: The following steps are involved: S1. Obtain the data object to be evaluated; S2, selecting an evaluation indicator from the evaluation indicator set; S3. According to the selected evaluation indicators, set constraint rules and quantify the evaluation indicators; S4. Conduct quantitative evaluation of the data object based on the quantified evaluation indicators to form an evaluation result.
2. A method for evaluating the quality of market supervision and law enforcement data as claimed in claim 1, characterized in that: The evaluation indicators in step S2 include: integrity indicator, consistency indicator, accuracy indicator, machine readability indicator, timeliness indicator, accessibility indicator and security indicator.
3. A method for evaluating the quality of market supervision and law enforcement data as claimed in claim 1, characterized in that: In step S3, the constraint rules include: empirical analysis method, missing value analysis method, logical relationship analysis method, repeated value analysis method, comparison analysis method, experimental observation method, phrase comparison analysis method and range analysis method.
4. A method for evaluating the quality of market supervision and law enforcement data as claimed in claim 1, characterized in that: In step S4, the quantitative evaluation of the data object includes: data set metadata evaluation, identification data record evaluation and content data record evaluation.
5. A method for evaluating the quality of market supervision and law enforcement data according to claim 1, characterized in that: In step S4, the quantitative evaluation method includes: a subjective evaluation method, a subjective and objective combination method, and an objective evaluation method.
6. A method for evaluating the quality of market supervision and law enforcement data as claimed in claim 5, characterized in that: The subjective evaluation methods include: expert meeting method, tomographic analysis method and Delphi method.
7. A method for evaluating the quality of market supervision and law enforcement data as claimed in claim 5, characterized in that: The subjective and objective combination method includes: Electre method, PROMETHEE method and fuzzy evaluation method.
8. A method for evaluating the quality of market supervision and law enforcement data as claimed in claim 5, characterized in that: The objective evaluation methods include: entropy weight method, coefficient of variation method and negative correlation coefficient method.
9. A storage medium, characterized in that: The storage medium stores instructions and data for implementing a market supervision and law enforcement data quality evaluation method as described in any one of claims 1 to 8.
10. A market supervision and law enforcement data quality evaluation device, characterized by: include: Processor and storage medium; the processor loads and executes instructions and data in the storage medium to implement a market supervision and law enforcement data quality evaluation method as described in any one of claims 1 to 8.
