A method, device and equipment for processing data across government systems based on RPA, and a storage medium

Abstract

The application discloses a kind of based on RPA's cross government affair system data processing method, device and equipment and storage medium, it is related to computer technical field, comprising: calling the field association relationship of each government affair system data specification information is obtained by preset form engine analysis, to determine the data to be processed;Call RPA execution cluster corresponding robot in each government affair system in the data to be processed to determine the merging weight of each field, to merge the data to be processed;Utilize Apriori algorithm to process the merging result to obtain association rule, carry out verification logic injection processing to obtain injection result, based on the device type of current device and injection result rendering current form interface;Call security gateway to verify current form interface, after verification, the sensitive field in current form interface is desensitized using ABAC model, to obtain target form interface, to process cross government affair system data, improve the efficiency of processing data.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to a cross-government system data processing method and device based on RPA, equipment and storage medium. BACKGROUND

[0002] At present, with the in-depth development of the information age, grassroots workers are facing unprecedented challenges. They need to deal with a large number of form filling tasks from many superior departments, which not only has a huge workload, but also seriously occupies the time applied to field research, thereby affecting the efficiency improvement of grassroots work. The current government informationization has the following pain points:

[0003] Data island problem: each department system is independently developed, the field naming, format and interface protocol are not unified, which leads to difficulty in data sharing.

[0004] Repeated labor: grassroots need to repeatedly enter the same data for different systems, with high error rate and repeated labor.

[0005] Synchronization delay: the existing ETL (Extract, Transform, Load, i.e. data extraction, transformation and loading process) tool relies on manual configuration of mapping relationship, cannot respond to business system updates in real time, and has high data synchronization delay.

[0006] Security risks: multiple systems store sensitive data in a decentralized manner, and the granularity of permission management is insufficient, which is prone to unauthorized access.

[0007] That is, existing solutions such as API (Application Programming Interface, i.e. application programming interface) integrated development have high cost, and traditional RPA (Robotic Process Automation, i.e. robotic process automation) only supports simple rule operations and cannot adapt to dynamic forms and complex verification logic in government scenarios.

[0008] From the above, how to improve the efficiency of processing data across government systems in the process of cross-government system data processing based on RPA is a problem to be solved at present. SUMMARY

[0009] Therefore, the purpose of the present application is to provide a cross-government system data processing method, device, equipment and storage medium based on RPA, which can improve the efficiency of processing data across government systems in the process of cross-government system data processing based on RPA. The specific scheme is as follows:

[0010] In a first aspect, the present application provides a cross-government system data processing method based on RPA, comprising:

[0011] The preset semantic analysis model in the preset form engine and the preset government domain ontology library are called to analyze data specification information of each government system, field association relationships across government systems are obtained, and the to-be-processed data is determined based on the field association relationships and the data specification information;

[0012] Robots corresponding to each government system in an RPA execution cluster are called, and a merging weight corresponding to each field in the to-be-processed data is determined based on a similarity clustering algorithm, a field use frequency of the to-be-processed data, government importance, and cross-government system coverage, the to-be-processed data is merged based on the merging weight, and a merging result is obtained;

[0013] The merging result is processed by using an Apriori algorithm, association rules are obtained, the merging result is subjected to a verification logic injection process by using the association rules, an injection result is obtained, a preset responsive layout technology, a preset resource preloading mechanism, and a preset government data platform interface are used, and a current form interface is rendered based on a device type of a current device and the injection result;

[0014] The current form interface is subjected to client format verification, server government rule verification, and cross-government system data verification by calling a security gateway, sensitive fields in the current form interface are desensitized by using an ABAC model after the verification passes, a target form interface is obtained, and cross-government system data is processed through the target form interface.

[0015] Optionally, before the preset semantic analysis model in the preset form engine and the preset government domain ontology library are called to analyze data specification information of each government system, field association relationships across government systems are obtained, and the to-be-processed data is determined based on the field association relationships and the data specification information, the method further includes the following steps.

[0016] It is determined whether each government system has a system interface, if the government system has a system interface, a system interface corresponding to each government system is identified, and field definition information corresponding to each government system is collected through the system interface; the field definition information includes a field name, a data type, and a constraint condition;

[0017] If the government system does not have a system interface, a non-intrusive data acquisition client is deployed in the government system, the data acquisition client is used to intercept query statements and operation traffic between an application program and a database of the government system, database information is obtained, reverse information generation is performed by using the database information, and data dictionary information is obtained;

[0018] generate to-be-processed data based on the field definition information and the data dictionary information, perform field labeling operations on each field of the to-be-processed data by using a preset visualization tool, and obtain data specification information; the fields include fields that have been encrypted and fields used for association across different government systems.

[0019] Optionally, the data specification information of each government system is analyzed by using a preset semantic analysis model in a preset form engine and a preset government domain ontology library to obtain a field association relationship across government systems, including:

[0020] A preset government domain ontology library covering government domains is constructed based on government information corresponding to each government domain; the government information includes government core concepts, government entity types, and government attribute relationships;

[0021] A preset semantic analysis model in a preset form engine is called to determine semantic similarity between field names of different government domains, an initial semantic determination result is obtained, and it is determined whether the initial semantic determination result is greater than a preset semantic threshold; if the initial semantic determination result is greater than the preset semantic threshold, it is determined that the government corresponding to each government domain corresponding to the initial semantic determination result is the same government, and the initial semantic determination result is set as a target semantic determination result;

[0022] The target semantic determination result is processed by using a government knowledge graph to obtain government context information, related standard information, and data value ranges corresponding to the target semantic determination result, so as to determine whether each field in the target semantic determination result has a mappable relationship based on the government context information, the related standard information, and the data value ranges; if each field has a mappable relationship, it is determined that there are multiple interpretations or to-be-processed fields with conflicts in the target semantic determination result, and the to-be-processed fields are processed to obtain a field association relationship across government systems.

[0023] Optionally, robots corresponding to each government system in an RPA execution cluster are called based on a similarity clustering algorithm, field use frequency of the to-be-processed data, government importance, and cross-government system coverage to determine a merging weight corresponding to each field in the to-be-processed data, and the to-be-processed data is merged based on the merging weight to obtain a merging result, including:

[0024] The field use frequency of the to-be-processed data, the government importance, and the cross-government system coverage are determined, and robots corresponding to each government system in an RPA execution cluster are called and the to-be-processed data is grouped based on a similarity clustering algorithm to obtain a grouping result;

[0025] perform extreme value smoothing on the to-be-processed data by using a preset logarithmic transformation algorithm and a frequency of field usage based on the fields in the grouping result, to obtain a smoothing result, and determine a first weight corresponding to the smoothing result;

[0026] determine a second weight corresponding to the to-be-processed data based on a preset importance level determination rule and a government affair importance of the to-be-processed data, and determine a third weight based on a cross-government system coverage of the to-be-processed data;

[0027] determine a weight score based on the first weight, the second weight and the third weight of each field in the grouping result, the frequency of field usage, the government affair importance and the cross-government system coverage, set a field with a weight score greater than a preset weight score threshold as a target field, and map the remaining fields in the grouping result to the target field to obtain a merging result.

[0028] Optionally, the merging result is processed by using the Apriori algorithm to obtain an association rule, the association rule is used for performing a verification logic injection processing on the merging result to obtain an injection result, a preset responsive layout technology, a preset resource preloading mechanism and a preset government affair data platform interface are used, and a current form interface is rendered based on a device type of a current device and the injection result, including:

[0029] an initial government affair rule is defined by using a preset structured configuration language; the initial government affair rule is used for specifying that when a to-be-judged field in a form meets a preset condition value, an associated field corresponding to the to-be-judged field is automatically displayed or hidden, and a data verification function corresponding to the associated field is activated or disabled;

[0030] a field combination satisfying a preset strong dependency relationship condition is obtained by analyzing historical form filling records by using a preset field combination determination algorithm, an association rule is determined based on the field combination and the merging result, the Apriori algorithm is used, and the association rule, the field combination and the initial government affair rule are used to determine a target government affair rule, the target government affair rule is used for performing a verification logic injection processing on the merging result to obtain an injection result;

[0031] a preset responsive layout technology and a preset virtual scrolling technology are used, and a current form interface is rendered and layout reorganized based on a device type of a current device and the injection result, a preset resource preloading mechanism is used for interface loading, to obtain an initial form interface, a voice navigation and a screen reader are integrated in the initial form interface through a preset government affair data platform interface, and a data prefill operation is performed, to obtain a current form interface.

[0032] Optionally, the security gateway is called to perform client-side format checking, server-side government rule checking, and cross-government system data checking on the current form interface, and after the checking passes, desensitization is performed on sensitive fields in the current form interface using an ABAC model to obtain a target form interface, so that cross-government system data is processed through the target form interface, including:

[0033] A preset regular expression pattern and a preset format checking function are used to perform format legality checking on data input by a user to the current form interface to obtain a format checking result, and a rule engine is deployed on a government service server to load and execute a government rule checking algorithm using the rule engine to obtain a government rule checking result; the government rule checking algorithm is used to perform logical consistency checking and government compliance review on the data;

[0034] If there is to-be-checked data across government systems in the current form interface, a smart contract deployed on a blockchain of the current government is called to access data sources in other government systems to perform comparison and checking to obtain a cross-government system data checking result;

[0035] If the format checking result, the government rule checking result, and the cross-government system data checking result all indicate that the checking passes, desensitization is performed on sensitive fields in the current form interface using an ABAC model to obtain a target form interface, so that cross-government system data is processed through the target form interface, and a processing operation is recorded in an immutable distributed ledger.

[0036] Optionally, after the target form interface is obtained by desensitizing the sensitive fields in the current form interface using the ABAC model, the method further includes:

[0037] A distributed version control mechanism is used to manage versions of the target form interface based on configuration information of the target form interface; each change corresponding to the target form interface corresponds to a unique version identifier;

[0038] When a change in a data field of a government system is monitored, an incremental training process of the ABAC model is triggered, and then a preset comparison test framework is used to test different form design versions corresponding to the target form interface to obtain a test result, so that parameters of the ABAC model are adjusted using the test result to obtain a new ABAC model;

[0039] According to a random allocation rule, different form design versions corresponding to the target form interface are allocated to different user groups, form filling results are obtained, and then a preset optimization algorithm is used to adjust the field arrangement order and field default value in the target form interface based on the form filling completion rate, form filling error rate and form filling time length of each form filling result, so as to obtain a new target form interface.

[0040] In a second aspect, the application provides an RPA-based cross-government system data processing device, comprising:

[0041] A field association relationship determination module is configured to call a preset semantic analysis model in a preset form engine and a preset government field ontology library to analyze data specification information of each government system, obtain a cross-government system field association relationship, and determine to-be-processed data based on the field association relationship and the data specification information.

[0042] A merging result determination module is configured to call robots corresponding to each government system in an RPA execution cluster and determine a merging weight corresponding to each field in the to-be-processed data based on a similarity clustering algorithm, field use frequency, government importance and cross-government system coverage rate, and merge the to-be-processed data based on the merging weight to obtain a merging result.

[0043] An injection result determination module is configured to process the merging result by using an Apriori algorithm to obtain an association rule, perform verification logic injection processing on the merging result by using the association rule to obtain an injection result, and render a current form interface based on a device type of a current device and the injection result by using a preset responsive layout technology, a preset resource preloading mechanism and a preset government data platform interface.

[0044] A form interface generation module is configured to call a security gateway to perform client-side format verification, server-side government rule verification and cross-government system data verification on the current form interface, and perform desensitization on sensitive fields in the current form interface by using an ABAC model after the verification is passed to obtain a target form interface, so as to process cross-government system data through the target form interface.

[0045] In a third aspect, the application provides an electronic device, comprising:

[0046] A memory is configured to save a computer program.

[0047] A processor is configured to execute the computer program to implement the aforementioned RPA-based cross-government system data processing method.

[0048] In a fourth aspect, the present application provides a computer readable storage medium for storing a computer program, wherein the computer program is executed by a processor to implement the foregoing RPA-based cross-government system data processing method.

[0049] As can be seen from the above, before performing the RPA-based cross-government system data processing, the preset semantic analysis model in the preset form engine and the preset government domain ontology library are called to analyze the data specification information of each government system, to obtain the field association relationship across the government systems, so as to determine the to-be-processed data based on the field association relationship and the data specification information; the robots corresponding to each government system in the RPA execution cluster are called and the similarity clustering algorithm, the field use frequency of the to-be-processed data, the government importance and the cross-government system coverage are used to determine the merging weight corresponding to each field in the to-be-processed data, so as to merge the to-be-processed data based on the merging weight to obtain a merging result; the Apriori algorithm is used to process the merging result to obtain an association rule, the association rule is used to perform logical injection processing on the merging result to obtain an injection result, then the preset responsive layout technology, the preset resource preloading mechanism and the preset government data platform interface are used, and the current form interface is rendered based on the device type of the current device and the injection result; the security gateway is called to perform client-side format verification, server-side government rule verification and cross-government system data verification on the current form interface, and after the verification is passed, the ABAC model is used to desensitize the sensitive fields in the current form interface to obtain a target form interface, so that the cross-government system data is processed through the target form interface.

[0050] It can be seen that the application first needs to call the preset semantic analysis model in the preset form engine and the preset government affair field ontology library to analyze the data specification information of each government affair system, obtain the field association relationship across the government affair systems, and determine the to-be-processed data based on the field association relationship and the data specification information; secondly, the robot corresponding to each government affair system in the RPA execution cluster is called and the merging weight corresponding to each field in the to-be-processed data is determined based on the similarity clustering algorithm, the field use frequency of the to-be-processed data, the government affair importance and the cross-government affair system coverage rate, so as to merge the to-be-processed data based on the merging weight and obtain the merging result; then, the Apriori algorithm is used to process the merging result, obtain the association rule, use the association rule to perform the check logic injection processing on the merging result, obtain the injection result, and then use the preset responsive layout technology, the preset resource preloading mechanism and the preset government affair data platform interface, and render the current form interface based on the device type of the current device and the injection result; finally, the security gateway is called to perform the client format check, the server side government affair rule check and the cross-government affair system data check on the current form interface, and after the check passes, the sensitive field in the current form interface is desensitized by using the ABAC model to obtain the target form interface, so as to process the cross-government affair system data through the target form interface. In this way, the efficiency of processing the data across the government affair systems is improved in the process of processing the data across the government affair systems based on RPA, and the user experience is improved. BRIEF DESCRIPTION OF DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only the embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on the provided drawings.

[0052] Figure 1 A flow chart of a specific RPA-based cross-government affair system data processing method disclosed by the present application;

[0053] Figure 2 A flow chart of a specific RPA-based cross-government affair system data processing method disclosed by the present application;

[0054] Figure 3 A structural schematic diagram of an RPA-based cross-government affair system data processing device disclosed by the present application;

[0055] Figure 4 A structural diagram of an electronic device disclosed by the present application. DETAILED DESCRIPTION

[0056] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described, obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work belong to the protection scope of the present application.

[0057] At present, with the in-depth development of the information age, grassroots workers are facing unprecedented challenges. They need to deal with a large number of form filling tasks from many superior departments, which not only has a huge workload, but also seriously occupies the time applied to field research, thereby affecting the efficiency improvement of grassroots work. Therefore, the present application provides a cross-government system data processing method based on RPA, which can improve the efficiency of processing data across government systems in the process of cross-government system data processing based on RPA.

[0058] Referring to Figure 1 The embodiment of the present application discloses a cross-government system data processing method based on RPA, comprising:

[0059] Step S11, calling a preset semantic analysis model in a preset form engine and a preset government domain ontology library to analyze data specification information of each government system, obtaining a field association relationship across government systems, to determine the to-be-processed data based on the field association relationship and the data specification information.

[0060] In the present embodiment, in the process of cross-government system data processing based on RPA, the specific process is as shown in Figure 2 Firstly, the present application utilizes a multi-modal dynamic form engine to dynamically generate scalable configured form components based on JSON Schema specification, and supports field-level, conditional rendering logic, for example: when the user selects "Enterprise Type" as "Limited Liability Company", the corresponding "Tax Registration Number" field is automatically triggered and displayed; and utilizes an RPA execution cluster containing special robots adapted to different systems (such as civil affairs robots, tax robots) and adopts a k8s (Kubernetes, an open source container orchestration management tool) deployment method to realize resource elastic scheduling. Furthermore, a data mapping hub integrated with a machine learning module is utilized to train and generate a cross-system field mapping model through historical operation logs. Subsequently, a secure gateway is called and end-to-end encryption is implemented on the transmission data based on the SM4 algorithm, and distributed evidence storage and tamper-proofing audit of operation logs are realized through blockchain technology.

[0061] In this embodiment, the data specification of multiple sources is first collected and semantically aligned, that is, the data specification information of each government affair system is collected by the triple way of OpenAPI Discovery (Open Application Programming Interface Discovery), database reverse engineering and manual annotation, then the field semantics is analyzed by using a multi-modal BERT model (Bidirectional Encoder Representations from Transformers), the naming ambiguity (such as "identity card number" and "certificate number" mapping) is eliminated in combination with a government affair ontology library, and a cross-system field association relationship is established. The system can automatically record the manual intervention decision to form a reusable mapping rule library and support incremental learning optimization.

[0062] It is worth mentioning that the access of heterogeneous data sources mainly includes the following modes:

[0063] Interface automatic sniffing: automatically identify the target system interface through the OpenAPI Discovery mechanism, and extract the field definition after analysis; database reverse engineering: for the legacy system without open interface, deploy a non-invasive client, intercept the database operation through the Hook technology, such as JDBC (Java Database Connectivity) traffic analysis, and reversely generate a data dictionary; manual supplement annotation: provide a visual tool for administrators to annotate special fields (such as encrypted field identification and cross-system associated fields).

[0064] Specifically, the preset semantic analysis model in the preset form engine and the preset government affair field ontology library are called to analyze the data specification information of each government affair system, to obtain the field association relationship across the government affair systems. Before determining the to-be-processed data based on the field association relationship and the data specification information, the method can further include: judging whether each government affair system has a system interface, if the government affair system has a system interface, identifying the system interface corresponding to each government affair system, to collect the field definition information corresponding to each government affair system through the system interface; the field definition information includes a field name, a data type and a constraint condition; if the government affair system does not have a system interface, performing a non-intrusive deployment operation of a data collection client in the government affair system, to perform interception processing of query statements and operation traffic between an application program and a database of the government affair system by using the data collection client, to obtain database information, and performing reverse information generation operation by using the database information, to obtain data dictionary information; generating the to-be-processed data based on the field definition information and the data dictionary information, and performing field labeling operation on each field of the to-be-processed data by using a preset visualization tool, to obtain the data specification information; the field includes a field that has been encrypted and a field used for association across different government affair systems.

[0065] Further, semantic alignment mainly includes the following ways: ontology modeling: constructing a government affair field ontology (Ontology), defining core concepts (such as "person" and "enterprise") and their attribute relationships; multi-modal semantic matching: text level: using a Sentence-BERT model to calculate the similarity of field names, such as "resident ID number" vs "ID number" → similarity 0.92; business level: associating the business meanings of fields (such as "family annual income" needs to be associated with the threshold of civil affairs subsistence allowances) through a knowledge graph; conflict resolution mechanism: when a field has ambiguity (such as "address" may refer to a household registration address or a current address), triggering an artificial confirmation process and recording decision rules.

[0066] Specifically, the preset semantic analysis model in the preset form engine and the preset government affair domain ontology library are called to analyze the data specification information of each government affair system to obtain the field association relationship across the government affair systems, which can include: a preset government affair domain ontology library covering the government affair domains is constructed based on the government affair information corresponding to each government affair domain; the government affair information includes government affair core concepts, government affair entity types, and government affair attribute relationships; a preset semantic analysis model in the preset form engine is called to determine the semantic similarity between the field names of different government affair domains to obtain an initial semantic determination result, and it is determined whether the initial semantic determination result is greater than a preset semantic threshold; if the initial semantic determination result is greater than the preset semantic threshold, it is determined that the government affairs corresponding to each government affair domain corresponding to the initial semantic determination result are the same government affair, and the initial semantic determination result is set as a target semantic determination result; the target semantic determination result is processed by using a government affair knowledge graph to obtain government affair context information, related standard information, and data value ranges corresponding to the target semantic determination result, so as to determine whether each field in the target semantic determination result has a mappable relationship based on the government affair context information, the related standard information, and the data value ranges; if each field has a mappable relationship, it is determined that there are multiple interpretations in the target semantic determination result or conflict fields to be processed in the government affair, and the fields to be processed are processed to obtain the field association relationship across the government affair systems.

[0067] In step S12, the robots corresponding to each of the government affair systems in the RPA execution cluster are called, and the merging weight corresponding to each field in the data to be processed is determined based on a similarity clustering algorithm, the field use frequency of the data to be processed, the government affair importance, and the cross-government affair system coverage rate, so as to merge the data to be processed based on the merging weight to obtain a merging result.

[0068] In this embodiment, the data to be processed needs to be merged, and in a specific implementation manner, the fields with a similarity greater than 0.85 are merged based on a DBSCAN clustering algorithm (Density-Based Spatial Clustering of Applications with Noise, i.e., a density-based spatial clustering algorithm with noise), and the merging weight is calculated by comprehensively using the frequency, the business criticality, and the cross-system coverage rate. Then, the explicit business rules (such as "enterprise type = foreign capital → display foreign exchange account field") are configured by using JSON (JavaScript Object Notation, i.e., JavaScript object notation), and the implicit association rules in the historical filling are mined by using an Apriori algorithm (a frequent item set-based association rule mining algorithm), so as to realize the dynamic display and hiding of the form fields and the injection of the verification logic.

[0069] It is worth mentioning that the intelligent field merging method is similarity clustering, that is, the DBSCAN algorithm is used to cluster the collected fields, and fields with a similarity > 0.85 are merged; the dynamic weight calculation rules are shown in Table 1:

[0070] Table 1 Schematic Table of Dynamic Weight Calculation Rules

[0071]

[0072] The final merging rules obtained are as follows:

[0073] ;

[0074] Among them, is the field usage frequency, is the business criticality, is the cross-system coverage rate. Subsequently, fields with a Score > 0.85 are retained.

[0075] Specifically, call the robots corresponding to each government affairs system in the RPA execution cluster and determine the merging weights corresponding to each field in the data to be processed based on the similarity clustering algorithm, the field usage frequency of the data to be processed, the government affairs importance, and the cross-government affairs system coverage rate, so as to merge the data to be processed based on the merging weights and obtain a merging result, which may include: determining the field usage frequency, government affairs importance, and cross-government affairs system coverage rate of the data to be processed, then calling the robots corresponding to each government affairs system in the RPA execution cluster and grouping the data to be processed based on the similarity clustering algorithm to obtain a grouping result; using a preset logarithmic transformation algorithm and performing extreme value smoothing processing on the data to be processed based on the field usage frequency in the grouping result to obtain a smoothing result, and determining the first weight corresponding to the smoothing result; using a preset importance level determination rule and determining the corresponding second weight based on the government affairs importance of the data to be processed, and determining the third weight determined based on the cross-government affairs system coverage rate of the data to be processed; determining a weight score based on each of the first weight, second weight, and third weight in the grouping result, as well as the field usage frequency, government affairs importance, and cross-government affairs system coverage rate, and setting the fields with a weight score greater than the preset weight score threshold as target fields, and mapping the remaining fields in the grouping result to the target fields to obtain a merging result.

[0076] It is worth mentioning that the dynamic display and hiding methods of form fields include explicit and implicit methods, where the explicit rules are configured as follows:

[0077] {

[0078] "when": "applicant_type == 'enterprise'",

[0079] "then": {

[0080] "show_fields": ["registered capital", "business license number"],

[0081] "hide_fields": ["family size"],

[0082] "validators": ["registered capital ≥ 10 million"]

[0083] }

[0084] }

[0085] And the implicit rules are configured as follows: analyze historical fill-in records by the Apriori algorithm to find associated field combinations (e.g., when "disability level = level 1" is selected, 95% of the time a disability certificate needs to be uploaded).

[0086] In step S13, the Apriori algorithm is used to process the merging result to obtain an association rule, the association rule is used to perform a check logic injection processing on the merging result to obtain an injection result, and then a preset responsive layout technology, a preset resource preloading mechanism, and a preset government affair data middle platform interface are used, and the current form interface is rendered based on the device type of the current device and the injection result.

[0087] In this embodiment, the embodiments of the present application can use CSS Grid (Cascading Style Sheets GridLayout Module, i.e., Cascading Style Sheets GridLayout Module) layout and virtual scrolling technology and automatically adapt PC / mobile interface according to device type, integrate voice navigation and screen reader compatible design, wherein the CSS Grid layout and virtual scrolling technology support second-level rendering of ten-thousand-level field forms, then predict the next operation through user behavior analysis and pre-load related field resources. Specifically, responsive layout is used and CSS Grid+Flexbox (i.e., a one-dimensional layout model) is used to realize dynamic layout reorganization to adapt to different screen sizes and meet the needs of multi-end rendering; at the same time, to improve efficiency and support intelligent filling, the embodiments of the present application support automatic completion of common information (such as permanent address) based on user historical data, and call government affair data middle platform interface to pre-fill authoritative data (such as enterprise credit code).

[0088] Specifically, the Apriori algorithm is used to process the merging result to obtain the association rule, the association rule is used for logical injection processing of the merging result, an injection result is obtained, and then a preset responsive layout technology, a preset resource preloading mechanism and a preset government data platform interface are used, and based on the device type of the current device and the injection result, the current form interface is rendered, which can include: defining an initial government rule by using a preset structured configuration language; the initial government rule is used to specify that when the to-be-judged field in the form meets the preset condition value, the associated field corresponding to the to-be-judged field is automatically displayed or hidden, and the data verification function corresponding to the associated field is activated or disabled; the field combination satisfying the preset strong dependency relationship condition is obtained by using a preset field combination determination algorithm to analyze the historical form filling records, and then the association rule is determined based on the field combination and the merging result, and then the target government rule is determined by using the Apriori algorithm and based on the association rule, the field combination and the initial government rule, and the target government rule is used for logical injection processing of the merging result to obtain the injection result; the current form interface is rendered and layout reorganization is performed based on the device type of the current device and the injection result by using the preset responsive layout technology and the preset virtual scrolling technology, and interface loading is performed by using the preset resource preloading mechanism to obtain an initial form interface, and then the voice navigation and the screen reader are integrated in the initial form interface through the preset government data platform interface and data prefilling operation is performed to obtain the current form interface.

[0089] In step S14, the security gateway is called to perform client format verification, server government rule verification and cross-government system data verification on the current form interface, and after the verification is passed, the sensitive fields in the current form interface are desensitized by using the ABAC model to obtain a target form interface, so that the cross-government system data is processed through the target form interface.

[0090] In this embodiment, the data dynamic verification and security control are selected, that is, a three-level verification system is implemented: front-end real-time format verification (regular expression), server business rule verification (Drools engine) and cross-system authoritative data verification (blockchain smart contract). Then, dynamic desensitization is performed on sensitive fields (such as displaying an ID number as 110**********1234), and the ABAC (attribute-based access control) model is used to control the field-level permission, all operations are transmitted by SM4 encryption and operation logs are recorded, and the multi-level verification system rules are as shown in Table 2:

[0091] Table 2 Multi-level verification system rule table

[0092]

[0093] Specifically, the security gateway is called to perform client-side format checking, server-side government affair rule checking and cross-government system data checking on the current form interface, and after the checking is passed, the ABAC model (Attribute-Based Access Control) is used to desensitize sensitive fields in the current form interface to obtain a target form interface, so as to process cross-government system data through the target form interface, which can include: using a preset regular expression pattern and a preset format checking function to perform format legality checking on data input by a user to the current form interface to obtain a format checking result, and deploying a rule engine on a government service server to load and execute a government rule checking algorithm using the rule engine to obtain a government rule checking result; the government rule checking algorithm is used to perform logical consistency checking and government compliance review on the data; if there is cross-government system data to be checked in the current form interface, a smart contract deployed on a blockchain of the current government is called to access data sources in other government systems for comparison and checking to obtain a cross-government system data checking result; if the format checking result, the government rule checking result and the cross-government system data checking result all indicate that the checking is passed, the ABAC model is used to desensitize sensitive fields in the current form interface to obtain a target form interface, so as to process cross-government system data through the target form interface, and record the processing operation to a tamper-proof distributed ledger.

[0094] That is, in order to ensure data security, the field-level permission control security enhancement measure is adopted in the embodiment of the application, and the ABAC model is adopted, for example:

[0095] IF user role = grassroots clerk AND business type = low-income application;

[0096] THEN editable fields = [family income, residence];

[0097] Dynamic desensitization: real-time addition of visible watermarks to sensitive fields (such as ID numbers).

[0098] Anti-tampering mechanism: use HMAC (keyed-Hash Message Authentication Code) to sign form data.

[0099] Subsequently, version management and intelligent evolution of the form are adopted, that is, Git-style version control, and rollback and difference comparison of form configuration are supported. The completion rate of different form layouts is evaluated through A / B testing (an experimental method for comparing the effects of different schemes), and Bayesian optimization is used to automatically adjust the field ordering. It is worth mentioning that when a business system field change is detected, the embodiment of the application can automatically trigger incremental training of the mapping model to ensure that the system continuously adapts to business changes and reduces maintenance costs.

[0100] In a specific embodiment, in terms of version control, the embodiments of the application adopt Git-style management, and a unique Commit ID is generated every time the form is modified, and version rollback and difference comparison are supported; when a field is deleted, the associated business system is automatically detected and a warning is issued. In use, continuous optimization is carried out, an A / B test framework is used, users are randomly assigned to use different form layouts, and indicators such as completion rate and error rate are collected for continuous optimization.

[0101] Specifically, after the check passes, the sensitive fields in the current form interface are desensitized using the ABAC model to obtain the target form interface. After processing the cross-government system data through the target form interface, it can further include: using a distributed version control mechanism and based on the configuration information of the target form interface, version management is performed on the target form interface; the changes corresponding to the target form interface each correspond to a unique version identifier; when it is monitored that the data field of the government system is changed, the incremental training process of the ABAC model is triggered, and then the different form design versions corresponding to the target form interface are tested using a preset comparison test framework to obtain test results, and the parameters of the ABAC model are adjusted using the test results to obtain a new ABAC model; different form design versions corresponding to the target form interface are allocated to different user groups according to a random allocation rule to obtain form filling results, and then a preset optimization algorithm is used and based on the form filling completion rate, the form filling error rate and the form filling duration of each form filling result, the field arrangement order and the field default value in the target form interface are adjusted to obtain a new target form interface.

[0102] In a specific embodiment, the specific implementation method in the embodiment of the application mainly includes four parts of data collection, form generation, automatic distribution, and security audit. Taking the enterprise opening a table as an example: first, data collection is performed, and the market supervision bureau API is parsed to obtain the fields ["unified social credit code", "enterprise name"]; second, the tax three system is inversely parsed to identify the fields ["taxpayer identification number", "registered address"]; then, the semantic engine is used to associate the "unified social credit code" and the "taxpayer identification number" (similarity 0.93). Further, the repeated fields are merged to generate a joint form ["unified code / tax number", "enterprise name", "registered address"], and then the business rules are injected: when the "enterprise type" is "foreign capital", the fields ["foreign exchange account", "investment country"] are displayed. Then, an RPA is used to submit a REST API request to the market supervision system, and for the environmental protection system whose interface is not open, keyboard input is simulated and OCR (Optical Character Recognition) verification is performed on the submission result. Finally, the hash value corresponding to the operation log ["timestamp: 2025-02-2014:23:15", "operator: Zhang San", "action: submit environmental approval"] is chained.

[0103] As can be seen from the above, the embodiment of the application first needs to call the preset semantic analysis model in the preset form engine and the preset government affair domain ontology library to analyze the data specification information of each government affair system, to obtain the field association relationship across government affair systems, so as to determine the to-be-processed data based on the field association relationship and the data specification information; secondly, the RPA execution cluster corresponding to each government affair system is called based on the similarity clustering algorithm, the field use frequency of the to-be-processed data, the government affair importance and the cross-government affair system coverage rate to determine the merging weight corresponding to each field in the to-be-processed data, so as to merge the to-be-processed data based on the merging weight to obtain a merging result; then, the Apriori algorithm is used to process the merging result to obtain an association rule, the association rule is used for logical injection processing of the merging result to obtain an injection result, and then the preset responsive layout technology, the preset resource preloading mechanism and the preset government affair data middleware interface are used, and the current form interface is rendered based on the device type of the current device and the injection result; finally, the security gateway is called to perform client format verification, server government affair rule verification and cross-government affair system data verification on the current form interface, and after the verification is passed, the sensitive fields in the current form interface are desensitized by using the ABAC model to obtain a target form interface, so that the cross-government affair system data is processed through the target form interface. In this way, the efficiency of processing the cross-government affair system data is improved in the process of processing the cross-government affair system data based on RPA, and the user experience is improved.

[0104] Correspondingly, referring toFigure 3 The application also provides an RPA-based cross-government system data processing device, which is shown as follows:

[0105] The field association relationship determination module 11 is configured to call a preset semantic analysis model in a preset form engine and a preset government field ontology library to analyze data specification information of each government system, so as to obtain a field association relationship across government systems, and determine the to-be-processed data based on the field association relationship and the data specification information.

[0106] The merging result determination module 12 is configured to call robots corresponding to each government system in an RPA execution cluster, and determine a merging weight corresponding to each field in the to-be-processed data based on a similarity clustering algorithm, a field use frequency of the to-be-processed data, government importance and cross-government system coverage, so as to merge the to-be-processed data based on the merging weight, and obtain a merging result.

[0107] The injection result determination module 13 is configured to process the merging result by using an Apriori algorithm, obtain an association rule, perform a verification logic injection process on the merging result by using the association rule, obtain an injection result, and then render a current form interface based on a device type of a current device and the injection result by using a preset responsive layout technology, a preset resource preloading mechanism and a preset government data platform interface.

[0108] The form interface generation module 14 is configured to perform client format verification, server government rule verification and cross-government system data verification on the current form interface by calling a security gateway, and perform desensitization on sensitive fields in the current form interface by using an ABAC model after the verification is passed, so as to obtain a target form interface, and process cross-government system data through the target form interface.

[0109] In some embodiments, the RPA-based cross-government system data processing device can further include:

[0110] The field definition information determination unit is configured to determine whether each government system has a system interface, identify a system interface corresponding to each government system if the government system has the system interface, and collect field definition information corresponding to each government system through the system interface; the field definition information includes a field name, a data type and a constraint condition.

[0111] The data dictionary information generation unit is configured to, if the government affair system does not have a system interface, perform a non-intrusive deployment operation of a data collection client in the government affair system, to obtain database information by using the data collection client to intercept query statements and operation traffic between an application and a database of the government affair system, and perform a reverse information generation operation by using the database information to obtain data dictionary information.

[0112] The data specification information determination unit is configured to generate to-be-processed data based on the field definition information and the data dictionary information, and perform field labeling operation on each field of the to-be-processed data by using a preset visualization tool to obtain data specification information; the fields include fields that have been encrypted and fields used for association across different government affair systems.

[0113] In some embodiments, the field association relationship determination module 11 can specifically include:

[0114] The government affair domain ontology library construction unit is configured to construct a preset government affair domain ontology library covering government affair domains based on government affair information corresponding to each government affair domain; the government affair information includes government affair core concepts, government affair entity types, and government affair attribute relationships.

[0115] The semantic determination result determination unit is configured to determine semantic similarity between field names of different government affair domains by calling a preset semantic analysis model in a preset form engine to obtain an initial semantic determination result, and determine whether the initial semantic determination result is greater than a preset semantic threshold; if the initial semantic determination result is greater than the preset semantic threshold, it is determined that government affairs corresponding to each government affair domain corresponding to the initial semantic determination result are the same government affair, and the initial semantic determination result is set as a target semantic determination result.

[0116] The to-be-processed field determination unit is configured to process the target semantic determination result by using a government affair knowledge graph to obtain government affair context information, related standard information, and data value ranges corresponding to the target semantic determination result, to determine whether each field in the target semantic determination result has a mappable relationship based on the government affair context information, the related standard information, and the data value ranges; if each field has a mappable relationship, it is determined that there are to-be-processed fields with multiple interpretations or conflicts in government affairs in the target semantic determination result, and the to-be-processed fields are processed to obtain a field association relationship across government affair systems.

[0117] In some embodiments, the merging result determination module 12 can specifically include:

[0118] The grouping result determination unit is configured to determine a field use frequency, a government affair importance, and a cross-government system coverage rate of the to-be-processed data, call robots corresponding to the government affair systems in an RPA execution cluster, and group the to-be-processed data based on a similarity clustering algorithm, to obtain a grouping result.

[0119] The first weight determination unit is configured to perform extreme value smoothing processing on the to-be-processed data based on the field use frequency in the grouping result by using a preset logarithmic transformation algorithm, to obtain a smoothing result, and determine a first weight corresponding to the smoothing result.

[0120] The second weight determination unit is configured to determine a second weight corresponding to the government affair importance of the to-be-processed data by using a preset importance level determination rule, and determine a third weight determined based on the cross-government system coverage rate of the to-be-processed data.

[0121] The merging result determination sub-unit is configured to determine a weight score based on the first weight, the second weight, and the third weight in the grouping result, and the field use frequency, the government affair importance, and the cross-government system coverage rate, set a field with a weight score greater than a preset weight score threshold as a target field, and map remaining fields in the grouping result to the target field, to obtain a merging result.

[0122] In some embodiments, the injection result determination module 13 can specifically include:

[0123] The associated field determination unit is configured to define an initial government affair rule by using a preset structured configuration language. The initial government affair rule is used to specify that, when a to-be-judged field in a form meets a preset condition value, an associated field corresponding to the to-be-judged field is automatically displayed or hidden, and a data verification function corresponding to the associated field is activated or disabled.

[0124] The injection result determination unit is configured to analyze historical form filling records by using a preset field combination determination algorithm, to obtain a field combination meeting a preset strong dependency relationship condition, determine an association rule based on the field combination and the merging result, determine a target government affair rule based on the association rule, the field combination, and the initial government affair rule by using an Apriori algorithm, and perform verification logic injection processing on the merging result by using the target government affair rule, to obtain an injection result.

[0125] The data pre-filling unit is configured to perform rendering and layout reorganization on the current form interface based on the device type of the current device and the injection result by using a preset responsive layout technology and a preset virtual scrolling technology, perform interface loading by using a preset resource preloading mechanism, obtain an initial form interface, integrate voice navigation and a screen reader in the initial form interface through a preset government data platform interface, and perform data pre-filling to obtain the current form interface.

[0126] In some embodiments, the form interface generation module 14 can specifically include:

[0127] The first verification result determination unit is configured to perform format legality checking on data input by a user to the current form interface by using a preset regular expression pattern and a preset format verification function, obtain a format verification result, and load and execute a government rule verification algorithm by using a rule engine deployed on a government service side to obtain a government rule verification result; the government rule verification algorithm is configured to perform logical consistency checking and government compliance review on the data.

[0128] The second verification result determination unit is configured to, if there is to-be-verified data across government systems in the current form interface, call a smart contract deployed on a blockchain of the current government to access data sources in other government systems for comparison and verification, and obtain a cross-government system data verification result.

[0129] The sensitive field desensitization unit is configured to, if the format verification result, the government rule verification result, and the cross-government system data verification result all indicate that the verification is passed, desensitize sensitive fields in the current form interface by using an ABAC model to obtain a target form interface, process cross-government system data through the target form interface, and record processing operations in a tamper-proof distributed ledger.

[0130] In some embodiments, the RPA-based cross-government system data processing apparatus can further include:

[0131] The version management unit is configured to perform version management on the target form interface by using a distributed version control mechanism and based on configuration information of the target form interface; each change corresponding to the target form interface corresponds to a unique version identifier.

[0132] The model parameter adjustment unit is configured to, when monitoring that a data field of a government system is changed, trigger an incremental training process of the ABAC model, test different form design versions corresponding to the target form interface by using a preset comparative test framework to obtain a test result, adjust parameters of the ABAC model by using the test result, and obtain a new ABAC model.

[0133] The form filling result determination unit is configured to allocate different form design versions corresponding to the target form interface to different user groups according to a random allocation rule, obtain form filling results, and then adjust the field arrangement order and the field default value in the target form interface by using a preset optimization algorithm and based on the form filling completion rate, the form filling error rate and the form filling time length of each form filling result, to obtain a new target form interface.

[0134] Further, the application further discloses an electronic device, Figure 4 is a structural diagram of an electronic device 20 according to an exemplary embodiment, and the content in the figure cannot be considered as any limitation on the use range of the application. The electronic device 20 can specifically include at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25 and a communication bus 26. The memory 22 is configured to store a computer program, and the processor 21 is configured to load and execute the computer program to implement the related steps in the RPA-based cross-government system data processing method disclosed in any of the preceding embodiments. In addition, the electronic device 20 in the embodiment can be an electronic computer.

[0135] In the embodiment, the power supply 23 is configured to provide working voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol followed by the communication interface 24 can be any communication protocol applicable to the technical solution of the application, which is not limited here; the input / output interface 25 is configured to obtain external input data or output data to the outside, and the specific interface type can be selected according to the specific application needs, which is not limited here.

[0136] In addition, the memory 22 as a carrier for resource storage can be a read-only memory, a random access memory, a magnetic disk or an optical disk, and the resources stored thereon can include an operating system 221, a computer program 222, etc., and the storage mode can be temporary storage or permanent storage.

[0137] The operating system 221 is configured to manage and control each hardware device on the electronic device 20 and the computer program 222, and can be Windows Server, Netware, Unix, Linux, etc. The computer program 222 can further include computer programs capable of completing other specific work in addition to the computer programs capable of completing the RPA-based cross-government system data processing method executed by the electronic device 20 disclosed in any of the preceding embodiments.

[0138] Further, the application also discloses a computer readable storage medium for storing a computer program, wherein the computer program is executed by a processor to realize the RPA-based cross-government system data processing method disclosed above. For the specific steps of the method, refer to the corresponding content disclosed in the foregoing embodiments, which will not be repeated here.

[0139] The various embodiments are described in the specification by progressive stages, and each embodiment focuses on the differences from other embodiments. For the device disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the relevant parts are referred to the method part.

[0140] The skilled person can further realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized by electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been described in the above description in general terms. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the application.

[0141] The steps of the method or algorithm described in combination with the embodiments disclosed herein can be directly implemented by hardware, a software module executed by a processor, or a combination of both. The software module can be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0142] Finally, it should be noted that in this document, relational terms such as first and second are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any such actual relationship or order between such entities or actions. Moreover, the terms "comprising", "including", or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or apparatus that includes a list of elements does not only include those elements, but also includes other elements not expressly listed or inherent to such process, method, article or apparatus. Without more limitations, an element defined by the statement "including a" does not exclude the presence of additional identical elements in the process, method, article or apparatus that includes the element.

[0143] The technical solutions provided by the present application are described in detail above, and the principles and implementation manners of the present application are described by using specific examples. The above description of the examples is only used to help understand the method of the present application and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manners and application ranges will be changed, and the above description of the content of the specification should not be understood as a limitation on the present application.

Claims

1. A method for processing data across government systems based on RPA, characterized in that, The method comprises the following steps: calling a preset semantic analysis model in a preset form engine and a preset government affair domain ontology library to analyze data specification information of each government affair system, obtaining a field association relationship across government affair systems, and determining to-be-processed data based on the field association relationship and the data specification information; wherein the process of calling the preset semantic analysis model in the preset form engine and the preset government affair domain ontology library to analyze the data specification information of each government affair system and obtaining the field association relationship across government affair systems comprises: collecting the data specification information through three modes of automatic sniffing through an interface, database reverse engineering and manual annotation, analyzing the field semantics of the data specification information by using a multi-modal BERT model, and eliminating naming ambiguity in the analysis result in combination with the preset government affair domain ontology library, so as to establish the field association relationship across government affair systems; calling a robot corresponding to each government affair system in an RPA execution cluster and determining a merging weight corresponding to each field in the to-be-processed data based on a similarity clustering algorithm, a field use frequency of the to-be-processed data, government affair importance and cross-government affair system coverage, so as to merge the to-be-processed data based on the merging weight and obtain a merging result; processing the merging result by using an Apriori algorithm to obtain an association rule, performing a check logic injection process on the merging result by using the association rule to obtain an injection result, and then rendering a current form interface based on a device type of a current device and the injection result by using a preset responsive layout technology, a preset resource preloading mechanism and a preset government affair data platform interface; calling a security gateway to perform client-side format checking, server-side government affair rule checking and cross-government affair system data checking on the current form interface, and performing desensitization on sensitive fields in the current form interface by using an ABAC model after the checking is passed to obtain a target form interface, so as to process cross-government affair system data through the target form interface; wherein the client-side format checking is performed by using a regular expression; the server-side government affair rule checking is performed by using a rule engine; and the cross-government affair system data checking is performed by using a blockchain smart contract.

2. The RPA-based cross-government system data processing method according to claim 1, characterized in that, Before the calling of the preset semantic analysis model in the preset form engine and the preset government affair domain ontology library to analyze the data specification information of each government affair system and the obtaining of the field association relationship across government affair systems based on the field association relationship and the data specification information, the method further comprises the following steps: determining whether each government affair system has a system interface, if the government affair system has a system interface, identifying a system interface corresponding to each government affair system, and collecting field definition information corresponding to each government affair system through the system interface; the field definition information comprises a field name, a data type and a constraint condition. If the government affair system does not have a system interface, a non-intrusive data collection client is deployed in the government affair system to intercept query statements and operation traffic between application programs and databases of the government affair system by using the data collection client, obtain database information, and generate reverse information by using the database information to obtain data dictionary information; Based on the field definition information and the data dictionary information, generate to-be-processed data, and perform field labeling on each field of the to-be-processed data by using a preset visualization tool to obtain data specification information; the fields include fields that have been encrypted and fields used for association across different government affair systems.

3. The RPA-based cross-government system data processing method according to claim 1, characterized in that, The preset semantic analysis model in the preset form engine is called to analyze the data specification information of each government affair system based on the preset semantic analysis model and the preset government affair domain ontology library to obtain a field association relationship across government affair systems, including: A preset government affair domain ontology library covering government affair domains is constructed based on government affair information corresponding to each government affair domain; the government affair information includes government affair core concepts, government affair entity types, and government affair attribute relationships; A preset semantic analysis model in a preset form engine is called to determine semantic similarity between field names of different government affair domains to obtain an initial semantic determination result, and it is judged whether the initial semantic determination result is greater than a preset semantic threshold; if the initial semantic determination result is greater than the preset semantic threshold, it is judged that the government affairs corresponding to each government affair domain corresponding to the initial semantic determination result are the same, and the initial semantic determination result is set as a target semantic determination result; The target semantic determination result is processed by using a government affair knowledge graph to obtain government affair context information, related standard information, and data value ranges corresponding to the target semantic determination result, so as to judge whether each field in the target semantic determination result has a mappable relationship based on the government affair context information, the related standard information, and the data value ranges; if each field has a mappable relationship, it is determined that there are multiple interpretations in the target semantic determination result or conflict to-be-processed fields in government affairs, and the to-be-processed fields are processed to obtain a field association relationship across government affair systems.

4. The RPA-based cross-government system data processing method according to claim 1, characterized in that, The robots corresponding to each government affair system in the RPA execution cluster are called based on a similarity clustering algorithm, a field use frequency of the to-be-processed data, government affair importance, and cross-government affair system coverage to determine a merging weight corresponding to each field in the to-be-processed data, and the to-be-processed data is merged based on the merging weight to obtain a merging result, including: The field use frequency, government affair importance, and cross-government affair system coverage of the to-be-processed data are determined, and then the robots corresponding to each government affair system in the RPA execution cluster are called and the to-be-processed data is grouped based on a similarity clustering algorithm to obtain a grouping result; The to-be-processed data is subjected to extreme value smoothing processing by using a preset logarithmic transformation algorithm based on the field use frequency in the grouping result to obtain a smoothing result, and a first weight corresponding to the smoothing result is determined. determine a second weight based on the government affair importance of the to-be-processed data according to a preset importance level determination rule, and determine a third weight based on a cross-government system coverage rate of the to-be-processed data; determine a weight score based on the first weight, the second weight and the third weight in the grouping result, the field use frequency, the government affair importance and the cross-government system coverage rate, set a field with a weight score greater than a preset weight score threshold as a target field, and map the remaining fields in the grouping result to the target field to obtain a merging result.

5. The RPA-based cross-government system data processing method according to claim 1, characterized in that, The Apriori algorithm is used to process the merging result to obtain an association rule, the association rule is used for logical injection processing of the merging result to obtain an injection result, and then a preset responsive layout technology, a preset resource preloading mechanism and a preset government data middle platform interface are used, and a current form interface is rendered based on a device type of a current device and the injection result, including: An initial government rule is defined by using a preset structured configuration language; the initial government rule is used to specify that when a to-be-judged field in a form meets a preset condition value, an associated field corresponding to the to-be-judged field is automatically displayed or hidden, and a data check function corresponding to the associated field is activated or disabled; A field combination satisfying a preset strong dependency condition is obtained by using a preset field combination determination algorithm to analyze historical form filling records, an association rule is determined based on the field combination and the merging result, and then a target government rule is determined by using the Apriori algorithm based on the association rule, the field combination and the initial government rule, so as to perform logical injection processing on the merging result by using the target government rule to obtain an injection result; An initial form interface is obtained by using a preset responsive layout technology and a preset virtual scrolling technology to render and reorganize a current form interface based on a device type of a current device and the injection result, and using a preset resource preloading mechanism to load an interface.

6. The RPA-based cross-government system data processing method according to claim 1, characterized in that, The security gateway is called to perform client format checking, server government rule checking and cross-government system data checking on the current form interface, and desensitization is performed on sensitive fields in the current form interface by using an ABAC model after the checking is passed, to obtain a target form interface, so as to process cross-government system data through the target form interface, including: A format checking result is obtained by using a preset regular expression pattern and a preset format checking function to perform format legality checking on data input by a user into a current form interface, and a rule engine is deployed on a government service side, so as to load and execute a government rule checking algorithm by using the rule engine to obtain a government rule checking result; the government rule checking algorithm is used for logical consistency checking and government compliance review of data; If there is to-be-verified data across government systems in the current form interface, a smart contract deployed on a blockchain of the current government is called to access data sources in other government systems for comparison and verification, and a cross-government system data verification result is obtained; If the format verification result, the government rule verification result, and the cross-government system data verification result all indicate that the verification is passed, the sensitive fields in the current form interface are desensitized using an ABAC model to obtain a target form interface, so that cross-government system data is processed through the target form interface, and processing operations are recorded in an immutable distributed ledger.

7. The RPA-based cross-government system data processing method according to any one of claims 1 to 6, characterized in that, After the sensitive fields in the current form interface are desensitized using the ABAC model to obtain the target form interface, the cross-government system data is processed through the target form interface, and the method further includes: The target form interface is version managed using a distributed version control mechanism and based on the configuration information of the target form interface; each change corresponding to the target form interface corresponds to a unique version identifier; When it is monitored that the data fields of the government system are changed, an incremental training process of the ABAC model is triggered, and then different form design versions corresponding to the target form interface are tested using a preset comparison test framework to obtain a test result, so that the parameters of the ABAC model are adjusted using the test result to obtain a new ABAC model; According to a random allocation rule, different form design versions corresponding to the target form interface are allocated to different user groups to obtain form filling results, and then a preset optimization algorithm is used to adjust the field arrangement order and field default value in the target form interface based on the form filling completion rate, the form filling error rate, and the form filling time length of each form filling result, to obtain a new target form interface.

8. A cross-government system data processing device based on RPA, characterized in that, It includes: A field association relationship determination module is configured to call a preset semantic analysis model in a preset form engine and a preset government domain ontology library to analyze data specification information of each government system, obtain a field association relationship across government systems, and determine to-be-processed data based on the field association relationship and the data specification information; wherein the process of calling the preset semantic analysis model in the preset form engine and the preset government domain ontology library to analyze the data specification information of each government system and obtain the field association relationship across government systems includes: collecting the data specification information through automatic sniffing, database reverse engineering, and manual annotation, analyzing the field semantics of the data specification information using a multi-modal BERT model, eliminating naming ambiguity in the analysis result in combination with the preset government domain ontology library, and establishing the field association relationship across government systems; A merging result determination module is configured to call robots corresponding to each government system in an RPA execution cluster and determine a merging weight corresponding to each field in the to-be-processed data based on a similarity clustering algorithm, a field use frequency of the to-be-processed data, government importance, and cross-government system coverage, and merge the to-be-processed data based on the merging weight to obtain a merging result; The injection result determination module is configured to process the merging result by using an Apriori algorithm to obtain an association rule, to perform a verification logic injection processing on the merging result by using the association rule to obtain an injection result, and to render a current form interface based on a device type of a current device and the injection result by using a preset responsive layout technology, a preset resource preloading mechanism, and a preset government affair data platform interface. The form interface generation module is configured to call a security gateway to perform a client format verification, a server government affair rule verification, and a cross-government system data verification on the current form interface, to perform desensitization on sensitive fields in the current form interface by using an ABAC model after the verifications pass, to obtain a target form interface, and to process cross-government system data by using the target form interface. The client format verification is performed by using a regular expression. The server government affair rule verification is performed by using a rule engine. The cross-government system data verification is performed by using a blockchain smart contract.

9. An electronic device, comprising: The computer program is stored in the memory and executed by the processor to implement the RPA-based cross-government system data processing method according to any one of claims 1 to 7. The computer program is stored in the memory and executed by the processor to implement the RPA-based cross-government system data processing method according to any one of claims 1 to 7. ​ 10. A computer-readable storage medium, characterized in that, ​

Images