A method for compliance testing of power companies' open data solutions using multi-dimensional fusion
By establishing a power grid data type graph and knowledge graph, the types and uses of open data from power companies are automatically identified, solving the problem of low efficiency in compliance testing of open data from power companies and achieving efficient compliance testing.
Patent Information
- Application Number
- CN202311262449.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-09-27
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2043-09-27
AI Technical Summary
In existing technologies, power companies are inefficient and prone to omissions when opening up data, making it impossible to conduct compliance checks efficiently.
By establishing a power grid data type map, using pinyin conversion, synonyms, and full-text search indexes, combined with knowledge graphs, data types are automatically identified, and audit requirements are adjusted according to usage and data volume to automatically generate test reports.
It enables efficient and compliant testing of open data from power companies, reducing manual retrieval and improving testing efficiency.
Smart Images

Figure CN117435729B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a method for compliance testing of open data schemes of power companies using multi-dimensional fusion, belonging to the field of data compliance assessment technology. Background Technology
[0002] With information technology increasingly covering all industries, there are more and more scenarios where government agencies and public institutions use power industry data for data analysis and decision-making. To ensure the legality and compliance of power data sharing, the first step is to conduct compliance testing on the sharing plan. Currently, power companies mainly rely on manual retrieval and comparison, checking each data item that needs to be shared before compiling a report, which is inefficient and prone to omissions. Summary of the Invention
[0003] The technical problem to be solved by this invention is to provide a method for compliance testing of open data schemes of power companies using multi-dimensional fusion, so as to overcome the shortcomings of the existing technology.
[0004] The technical solution of this invention is: a method for compliance testing of open data schemes of power companies using multi-dimensional fusion, comprising the following steps:
[0005] S1. Construct a power grid data type map Gm based on the guidelines for classification and grading of power data security;
[0006] S2. For each root word, establish recognition rules, identify the rules that the string satisfies, and obtain the data type corresponding to the rules;
[0007] S3. Upload a list of data items, and use recognition rules and knowledge graphs to automatically identify the data type information in the uploaded data list;
[0008] S4. Retrieve the list of data items, automatically match each data item, and obtain its business type, hierarchical information, anonymization requirements, and approval requirements;
[0009] S5. Match the open uses in the plan. If the desensitization requirements are met, delete the desensitization requirements found in step S4. Otherwise, retain the desensitization requirements.
[0010] S6. Based on the amount of data in the plan, automatically adjust the audit requirements retrieved in step S4 of the plan. If the amount of data is greater than 100,000, then a mandatory audit will be conducted.
[0011] S7. Based on the estimated number of entities in the plan, automatically adjust the review requirements queried in step S4 of the plan. If the number of dimensions is greater than 100,000, a review will be forced.
[0012] S8. The test results will be automatically generated into a test report, thus completing the test.
[0013] Specifically, step S1 includes:
[0014] A basic graph Gm is created using a power grid data classification model. The name of the data item is used as the node name, the name of the data item is used as the root word, and the data type is used as the node label.
[0015] For each node Nm, perform pinyin conversion, take the full pinyin of the name and the first letter of the pinyin as nodes, and connect them to Nm;
[0016] Manually add synonyms for nodes;
[0017] A full-text search index is built for all nodes, and based on the full-text search index, a graph data recognition interface GInterface based on text similarity comparison is provided. The graph data recognition interface GInterface returns a set C of all data node names with similarity higher than a set value, and sorts them from highest to lowest similarity, with the first element having the highest similarity.
[0018] Traverse all metadata in the power data asset catalog, use the name of the metadata as the retrieval parameter, input it into the graph data recognition interface GInterface, perform data recognition, if the recognition result C has data, take the first element C1, create a node for the metadata and connect it to C1, and create an index for the node at the same time;
[0019] For metadata that cannot be identified, it is either manually associated with an existing node, or Step 1 is repeated to create a new metadata root, create an index, and then use the root to traverse the asset catalog once to find metadata with similar text, and create new nodes and connections.
[0020] Furthermore, step S1 also includes:
[0021] Call the Baidu Translate API to connect all the English translations of node Nm as nodes to Nm.
[0022] Specifically, step S3 includes:
[0023] Upload an Excel spreadsheet containing metadata to be processed;
[0024] Create an array A, whose length is equal to the number of sheets in the metadata Excel table. Each element of A is a Map, where the key is a string and the value is an array of Objects. The metadata Excel table provides the system with a list of uploaded data items.
[0025] Starting from the first sheet, iterate through all sheets. First, get the position i of the sheet in the metadata Excel table, and create new arrays ListH and ListV.
[0026] Iterate through A starting from the first element and extract the element M.
[0027] Specifically, the method of iterating through all sheets starting from the first sheet, first obtaining the sheet's position i in the metadata Excel table, and then creating new arrays ListH and ListV, is as follows:
[0028] Iterate through each column in the sheet, labeling the position as ir. Extract the first 30 rows of data as a collection ListF. If there are fewer than 30 rows, extract all rows and place ListF at the ir-th position of ListV.
[0029] Read the first row of ListF and put it into the ir position in ListH. If the first row is empty, put a NULL object in it.
[0030] After traversing a sheet, retrieve the i-th object M from array A, put ListH into M with the key equal to header, and put ListV into M with the key equal to value.
[0031] Specifically, the method for traversing A starting from its first element and retrieving element M is as follows:
[0032] Retrieve the array ListH from array A where the key equals header; iterate through the ListH array, taking each non-NULL element as a parameter, inputting it into the graph data recognition interface GInterface, and obtaining the node set Cg. If Cg is not empty, put Cg into M;
[0033] Take out the array ListV in the prime group A where the key equals the value. Iterate through ListV starting from the first element. Each element of ListV is denoted as array ListF. Iterate through ListF. For each non-empty element in ListF, execute the rule-based recognition interface RInterface to obtain the data result set Cr. If Cr is not empty, put Cr into M with the key equal to rule.
[0034] If Cr is not empty, take the first element of Cr and put it into M, with the key equal to dataType, which is the basic data type of this column;
[0035] If Cr is empty and Cg is not empty, take the first element of Cg and put it into M, with the key equal to dataType, as the basic data type of this column;
[0036] If Cr and Cg are both empty, then no basic data type is set.
[0037] Furthermore, it also includes displaying the results of Cg and Cr on the system interface.
[0038] Specifically, the recognition rule is a regular expression written in JavaScript.
[0039] The beneficial effects of this invention are as follows: Compared with the prior art, this invention first establishes a power grid data type map (Gm) based on the power data security classification and grading guidelines. This map includes data item names as nodes, data types as node labels, and also involves pinyin conversion, adding synonyms for nodes, full-text search indexing, and a graph data recognition interface based on text similarity. Recognition rules are established for each root word to identify the rules that strings satisfy and obtain the corresponding data type. The uploaded data item list file will automatically identify the data type information of the data items using these recognition rules and the knowledge graph. Retrieval is performed through the data item list to automatically match each data item and obtain its business type, grading information, anonymization requirements, and approval requirements. Anonymization is performed based on the open purpose: matching is performed according to the open purpose in the plan; if the anonymization requirements are met, the retrieved anonymization requirements are deleted; otherwise, the anonymization requirements are retained. The retrieved approval requirements are automatically adjusted based on the data volume and estimated number of subjects in the plan. If the data volume or dimension volume exceeds 100,000, a mandatory approval is enforced. Finally, all detection results are automatically generated into a detection report, completing the entire detection process. The entire process mainly relies on machine retrieval and comparison, resulting in higher efficiency for compliance testing. Attached Figure Description
[0040] Figure 1 This is the basic data item type of the present invention;
[0041] Figure 2 The data items in this invention are represented by their Pinyin and English nodes;
[0042] Figure 3 Synonyms manually added for this invention;
[0043] Figure 4 This is a knowledge graph of power metadata types for the present invention;
[0044] Figure 5 This is a diagram of the interface for uploading files to be tested according to the present invention;
[0045] Figure 6 This is the interface for filling in detailed information about the present invention.
[0046] Figure 7 This is a flowchart illustrating the compliance testing process for the data openness scheme of this invention.
[0047] Figure 8 This is an example of the text recognition process of the present invention. Detailed Implementation
[0048] The technical problem this invention aims to solve is to provide a method for compliance checks on power companies' open data schemes using multi-dimensional fusion. This method automatically detects the data to be opened from multiple dimensions, including data type, data classification, data volume, dimensionality, and intended use, and automatically generates a detection report. This reduces the need for manual retrieval, comparison, and judgment, thereby improving the efficiency of compliance checks.
[0049] The data sharing scheme described in this invention mainly includes five fields: data item list, open object, open purpose, data volume, and estimated number of subjects.
[0050] A data item list is a collection of data items. Data items are metadata fields that need to be made public, such as: homeowner's name, household number, monthly electricity consumption, etc.
[0051] Open objects refer to data export object units.
[0052] The scope of use is open, including three options: routine, audit, and overseas.
[0053] Data volume refers to the maximum number of data records that can be opened at one time. As shown in the table below, the data volume is 5, which means that a total of 5 data records were opened.
[0054] Table 1 Open Data Content Table
[0055] Head of household ID card Account Number Monthly electricity consumption Yang X 520xxxxxxxxxxxxxx1 2xxxxxx9 20 Zhang X 520xxxxxxxxxxxxxx2 7xxxxxx3 20 Wang X 520xxxxxxxxxxxxxx3 3xxxxxx8 20 Li X 520xxxxxxxxxxxxxx4 4xxxxxx9 20 Yang B 520xxxxxxxxxxxxxx5 7xxxxxx9 20
[0056] The estimated number of subjects refers to the number of data items that are grouped and counted at one time. As shown in Table 1, if the data is counted based on the ID card, the data volume is 4, which means that the data of 4 people are open.
[0057] Example 1:
[0058] This embodiment includes the following steps:
[0059] 1. Based on the guidelines for the classification and grading of power data security, construct a power grid data type map Gm. The construction approach is as follows:
[0060] ① Create a power grid data type graph Gm. Use a power grid data classification model to create a basic graph Gm. The name of the data item serves as the node name, becoming the "root word," and the data type serves as the node label, such as... Figure 1 As shown.
[0061] ② Perform pinyin conversion on each node Nm, using the full pinyin spelling and the first letter of the pinyin as nodes, and connect them to Nm; call the Baidu Translate API to use all the English translations of node Nm as nodes, and connect them to Nm, such as... Figure 2 As shown.
[0062] ③ Manually add synonyms for nodes, such as synonyms for name: name, title, etc. Figure 3 As shown.
[0063] ④ Build a full-text search index for all nodes, and based on the full-text search index, provide a graph data recognition interface GInterface based on text similarity comparison. This interface returns a set C of all data node names with similarity higher than a set value, and sorts them from highest to lowest similarity, with the first element having the highest similarity.
[0064] ⑤ Traverse all metadata in the power data asset catalog, using the metadata name as the retrieval parameter, and input it into the graph data recognition interface GInterface provided in step 4 for data recognition. If the recognition result C contains data, then take the first element C1, create a node for that metadata, connect it to C1, and simultaneously create an index for that node, as follows. Figure 4 As shown.
[0065] ⑥ For metadata that cannot be identified, manually associate it with an existing node, or repeat step 1 to create a new metadata root, build an index, and then use the root to traverse the asset catalog once to find metadata with similar text, and build new nodes and connections.
[0066] 2. Establish recognition rules for each word root, using regular expressions written in JavaScript.
[0067] For example, the rules for recognizing names are as follows:
[0068] function execSb(str){
[0069] var reg= / ^[Zhao,Qian,Sun,Li,Zhou,Wu,Zheng,Wang,Feng,Chen,Chu,Wei,Jiang,Shen,Han,Yang,Zhu,Qin,You,Xu,He,Lü,Shi,Zhang,Kong,Cao,Yan,Hua,Jin,Wei,Tao,Jiang,Qi,Xie,Zou,Yu,Bai,Shui,Dou,Zhang,Yun,Su,Pan,Ge,Xi,Fan,Peng,Lang,Lu,Wei,Chang,Ma,Miao,Feng,Hua,Fang,Yuan,Ren,Yuan] ,Liu,Feng,Bao,Shi,Tang,Fei,Lian,Cen,Xue,Lei,He,Ni,Tang,Teng,Yin,Luo,Bi,Hao,Wu,An,Chang,Le,Yu,Shi,Fu,Pi,Bian,Qi,Kang,Wu,Yu,Yuan,Bu,Gu,Meng,Ping,Huang,He,Mu,Xiao,Yin,Yao,Shao,Zhan,Wang,Qi,Mao,Yu,Di,Mi,Bei,Ming,Zang,Ji,Fu,Cheng,Dai,Tan,Song,Mao,Pang,Xiong Ji, Shu, Qu, Xiang, Zhu, Dong, Liang, Du, Ruan, Lan, Min, Xi, Ji, Ma, Qiang, Jia, Lu, Lou, Wei, Jiang, Tong, Yan, Guo, Mei, Sheng, Lin, Diao, Zhong, Xu, Qiu, Luo, Gao, Xia, Cai, Tian, Fan, Hu, Ling, Huo, Yu, Wan, Zhi, Ke, Zan, Guan, Lu, Mo, Jing, Fang, Qiu, Miao, Gan, Jie, Ying, Zong, Ding, Xuan, Ben, Deng, Yu, Shan, Hang Hong, Bao, Zhu, Zuo, Shi, Cui, Ji, Niu, Gong, Cheng, Ji, Xing, Hua, Pei, Lu, Rong, Weng, Xun, Yang, Yu, Hui, Zhen, Qu, Jia, Feng, Rui, Yi, Chu, Jin, Ji, Bing, Mi, Song, Jing, Duan, Fu, Wu, Wu, Jiao, Ba, Gong, Mu, Wei, Shan, Gu, Che, Hou, Mi, Peng, Quan, Xi, Ban, Yang, Qiu, Zhong, Yi, Gong, Ning, Qiu, Luan, Bao, Gan, Xie ,Li,Rong,Zu,Wu,Fu,Liu,Jing,Zhan,Shu,Long,Ye,Xing,Si,Shao,Gao,Li,Ji,Bo,Yin,Su,Bai,Huai,Pu,Tai,Cong,E,Suo,Xian,Ji,Lai,Zhuo,Lin,Tu,Meng,Chi,Qiao,Yin,Yu,Xu,Neng,Cang,Shuang,Wen,Xin,Dang,Zhai,Tan,Gong,Lao,Pang,Ji,Shen,Fu,Du,Ran,Zai,Li,Yong,Xi,Qu,Sang,Gui ,Pu,Niu,Shou,Tong,Bian,Hu,Yan,Ji,Jia,Pu,Shang,Nong,Wen,Bie,Zhuang,Yan,Chai,Qu,Yan,Chong,Mu,Lian,Ru,Xi,Huan,Ai,Yu,Rong,Xiang,Gu,Yi,Shen,Ge,Liao,Yu,Zhong,Ji,Ju,Heng,Bu,Du,Geng,Man,Hong,Kang,Guo,Wen,Kou,Guang,Lu,Que,Dong,Ou,Shu,Wo,Li,Wei,Yue,Kui,Long,Shi,Gong She, Nie, Chao, Gou, Ao, Rong, Leng, Zi, Xin, Kan, Na, Jian, Rao, Kong, Zeng, Wu, Sha, Nie, Yang, Ju, Xu, Feng, Chao, Guan, Kuai, Xiang, Cha, Hou, Jing, Hong, You, Zhu, Quan, Qiu, Gai, Yi, Huan, Gong, Wanqi, Sima, Shangguan, Ouyang, Xiahou, Zhuge, Wenren, Dongfang, Helian, Huangfu, Yuchi, Gongyang, Tantai, Gongye, Zongzheng, Puyang Chunyu, Shanyu, Taishu, Shentu, Gongsun, Zhongsun, Xuanyuan, Linghu, Zhongli, Yuwen, Changsun, Murong, Xianyu, Lüqiu, Situ, Sikong, Buguan, Sikou, Zhang, Du, Ziche, Zhuansun, Duanmu, Wuma, Gongxi, Qidiao, Lezheng, Rangsi, Gongliang, Tuoba, Jiagu, Zaifu, Guliang, Jin, Chu, Yan, Fa, Ru, Yan, Tu, Qin, Duangan, Baili, Dongguo, NanmenHuyan, Gui, Hai, Yangshe, Weisheng, Yue, Shuai, Gou, Kang, Kuang, Hou, You, Qin, Liangqiu, Zuoqiu, Dongmen, Ximen, Shang, Mou, She, Er, Bo, Shang, Nangong, Mo, Ha, Qiao, Da, Nian, Ai, Yang, Tong, Diwu, Yan, Fu [\\u4e00-\\u9fa5]{1,4}$ / ;,
[0070] return reg.test(str);
[0071] }
[0072] Then, a rule-based recognition interface RInterface is provided, with a string as the parameter. Based on the configured JS rules, the interface identifies the rules that the string satisfies, thereby obtaining the data type corresponding to the rule.
[0073] 3. Build an information system, developed using Java. The system should provide the function of uploading a list of data items (Excel format). Figure 5 As shown, for the uploaded data list, the system will automatically identify the data type information using recognition rules and knowledge graphs, and detect compliance processes such as... Figure 7 As shown in Table 2, the uploaded information can contain multiple sheets (tables in Excel), and the data for one of the sheets is shown in Table 2:
[0074] Table 2 Open Data Content Table
[0075]
[0076] The system identification method is as follows:
[0077] ① Upload a metadata Excel sheet to be processed; create an array A, the length of which is equal to the number of sheets in the metadata Excel sheet. Each element of A is a Map, where the key is a string and the value is an Object array. The metadata Excel sheet provides the system with a list of uploaded data items.
[0078] ② Starting from the first sheet, iterate through all sheets. First, get the position i of the sheet in the metadata Excel table, and create new arrays ListH and ListV.
[0079] Iterate through each column in the sheet, labeling its position as ir. Extract the first 30 rows of data and create a collection ListF. If there are fewer than 30 rows, extract all rows and place ListF at the ir-th position of ListV.
[0080] Read the first row of ListF and put it into the ir position in ListH. If the first row is empty, put a NULL object into it.
[0081] After traversing a sheet, retrieve the i-th object M from array A, put ListH into M with the key equal to header, and put ListV into M with the key equal to value.
[0082] ③ Start traversing A from its first element and extract element M.
[0083] Retrieve the array ListH from array A where the key equals header; iterate through the ListH array, taking each non-NULL element as a parameter, inputting it into the graph data recognition interface GInterface to obtain the node set Cg. If Cg is not empty, put Cg into M. Refer to the recognition process... Figure 8 ;
[0084] Retrieve ListV from array A, where each key equals its value. Iterate through ListV starting from the first element. Let each element of ListV be an array ListF. Iterate through ListF, and for each non-empty element in ListF, execute the rule-based identification and obtain the data result set Cr. If Cr is not empty...
[0085] Then put Cr into M;
[0086] If Cr is not empty, take the first element of Cr and put it into M, with the key equal to dataType, which serves as the basic data type for this column.
[0087] If Cr is empty and Cg is not empty, take the first element of Cg and put it into M, with the key equal to dataType, which serves as the basic data type for this column.
[0088] If both Cr and Cg are empty, no basic data type is set. This completes the automatic data type identification process. The results for Cg and Cr can then be displayed on the system interface, allowing users to manually adjust and replace the basic data type.
[0089] 4. The system retrieves the data item list in the plan through the Southern Power Grid Data Asset Management System, automatically matches each data item, and obtains its business type, hierarchical information, anonymization requirements, and approval requirements.
[0090] 5. The system requires users to fill in detailed solution information, such as... Figure 6 As shown. Matching is performed based on the open use in the plan. If it is a type such as law enforcement or auditing, the de-identification requirements found in step 4 are deleted; otherwise, the de-identification requirements are retained.
[0091] 6. The system will automatically adjust the audit requirements retrieved in step 4 of the plan based on the amount of data in the plan. If the amount of data exceeds 100,000, an audit will be mandatory.
[0092] 7. The system automatically adjusts the review requirements queried in step 2 of the plan based on the estimated number of entities in the plan. If the number of dimensions is greater than 100,000, a review will be forced.
[0093] 8. The test results will be automatically generated into a test report, thus completing the testing process.
[0094] All aspects not detailed herein are well-known to those skilled in the art. Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of this invention and not intended to limit it. Although the invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of this invention without departing from the spirit and scope of the invention, and all such modifications and substitutions should be covered within the scope of the claims of this invention.
Claims
1. A method for compliance testing of power companies' open data solutions using multi-dimensional fusion, characterized in that, The process includes the following steps: S1. Construct a power grid data type map Gm based on the guidelines for classification and grading power data security; S2. Establish recognition rules for each root word, identify the rules that a string satisfies, and obtain the data type corresponding to the rule; S3. Upload a data item list file, and use the recognition rules and knowledge graph to automatically identify the uploaded data list and determine the data type information within it. S4. Retrieve the list of data items, automatically match each data item, and obtain its business type, hierarchical information, anonymization requirements, and approval requirements; S5. Match the open uses in the plan. If the de-identification requirements are met, delete the de-identification requirements found in step S4; otherwise, retain the de-identification requirements. S6. Automatically adjust the review requirements found in step S4 of the plan based on the data volume. If the data volume is greater than 100,000, a review will be mandatory. S7. Automatically adjust the review requirements found in step S4 of the plan based on the estimated number of entities. If the number of dimensions is greater than 100,000, a review will be mandatory. S8. Automatically generate a test report from the test results to complete the test work; Step S1 specifically includes: using the power grid data classification model to create a basic graph Gm, where the name of the data item is used as the node name, the name of the data item is used as the root word, and the data type is used as the node label; For each node Nm, perform pinyin conversion, using the full pinyin and the first letter of the pinyin as nodes, and connect them to Nm; manually add synonyms for nodes; build a full-text search index for all nodes, and based on the full-text search index, provide a graph data recognition interface GInterface based on text similarity comparison. The graph data recognition interface GInterface returns a set C of all data node names with similarity higher than a set value, sorted from highest to lowest similarity, with the first element having the highest similarity; traverse all metadata in the power data asset catalog, using the metadata name as a search parameter, inputting it into the graph data recognition interface GInterface for data recognition. If the recognition result C contains data, take the first element C1, create a node for that metadata, connect it to C1, and simultaneously create an index for that node; for metadata that cannot be recognized, manually associate it with an existing node, or repeat step 1, create a new metadata root, build an index, and then use this root to traverse the asset catalog once, searching for text-similar metadata, and creating new nodes and connections.
2. The method for compliance testing of power company open data schemes using multi-dimensional fusion according to claim 1, characterized in that, Step S1 further includes: calling the Baidu Translate API to connect all the English translations of node Nm as nodes to Nm.
3. The method for compliance testing of power company open data schemes using multi-dimensional fusion according to claim 1, characterized in that, Step S3 specifically includes: Uploading a metadata Excel sheet to be processed, the metadata Excel sheet providing the system with a list of uploaded data items; Creating an array A, the length of which is equal to the number of sheets in the metadata Excel sheet, each element of A being a Map, where the key is a string and the value is an Object array; Traversing all sheets starting from the first sheet, first obtaining the position i of the sheet in the metadata Excel sheet, creating new arrays ListH and ListV; Traversing A starting from the first element of A, retrieving element M.
4. The method for compliance testing of power company open data schemes using multi-dimensional fusion according to claim 3, characterized in that, The method for iterating through all sheets starting from the first sheet is as follows: First, obtain the position i of the sheet in the metadata Excel table. Create new arrays ListH and ListV. Specifically, iterate through each column in the sheet, marking the position as ir. Extract the first 30 rows of data as a collection ListF. If there are not 30 rows, extract all rows and place ListF at the ir-th position in ListV. Read the first row of ListF and place it at the ir-th position in ListH. If the first row is empty, place a NULL object. After completing the iteration of one sheet, extract the i-th object M from array A, insert ListH into M with the key equal to header, and insert ListV into M with the key equal to value.
5. The method for compliance testing of power company open data schemes using multi-dimensional fusion according to claim 3, characterized in that, The specific method for traversing A from its first element and retrieving element M is as follows: Retrieve the array ListH from array A where the key equals header; traverse the ListH array, taking each non-NULL element as a parameter and inputting it into the graph data recognition interface GInterface to obtain the node set Cg. If Cg is not empty, place Cg into M; Retrieve the array ListV from array A where the key equals value, traversing ListV from its first element. Each element of ListV is denoted as array ListF. Traverse ListF, and for each non-empty element in ListF, execute the rule-based recognition interface RInterface to obtain the data result set Cr. If Cr is not empty, place Cr into M; If Cr is not empty, take the first element of Cr and place it into M, with the key equal to dataType, as the basic data type of this column; If Cr is empty and Cg is not empty, take the first element of Cg and place it into M, with the key equal to dataType, as the basic data type of this column; If Cr and Cg are empty... If all values are empty, then no basic data type is set.
6. The method for compliance testing of power company open data schemes using multi-dimensional fusion according to claim 5, characterized in that, It also includes displaying the results of Cg and Cr on the system interface.
7. The method for compliance testing of power company open data schemes using multi-dimensional fusion according to claim 1, characterized in that, The recognition rules are regular expressions written in JavaScript.
Citation Information
Patent Citations
Private data use compliance check method based on knowledge graph
CN115906852A
