Report automatic generation method, device and equipment and storage medium

Abstract

The present application relates to the technical field of data processing, and discloses a report automatic generation method, device, equipment and computer readable storage medium, the method comprises: obtaining historical report and sending timestamp, and the latest reporting rule and publishing timestamp; when the publishing timestamp is later than the sending timestamp, the latest historical reporting rule is obtained, the difference between the rule elements corresponding to the latest reporting rule and the historical reporting rule is obtained by comparing the latest reporting rule, and a template modification operation is obtained; a template modification instruction is generated according to the template modification operation, the field unit in the report template is automatically modified by using the template modification instruction, and a target template is generated; according to the latest reporting rule and the target template, the reporting data meeting the requirements of the latest reporting rule is extracted from the standard data, the reporting data is filled into the corresponding position of the target template to generate a target report. The present application can be applied to the business system platform of financial technology, medical health and the like, and can automatically generate a target report according to the reporting rule change.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method, apparatus, device, and storage medium for automatically generating reports. Background Technology

[0002] In fields such as healthcare and fintech, institutions are required to regularly generate and submit a large number of structured reports. For example, medical institutions need to report disease surveillance, medical insurance settlement, or DRG / DIP grouping data as required by health departments; financial institutions need to submit compliance reports such as capital adequacy ratio and liquidity coverage ratio to regulatory agencies or group headquarters. These reports are highly dependent on pre-defined reporting rules and fixed templates, with the rules clearly defining field definitions, calculation logic, validation thresholds, and format specifications.

[0003] However, the reporting rules in these areas are updated extremely frequently: adjustments to medical policies often lead to changes in disease coding, cost-sharing rules, or data granularity; new financial regulations also continuously adjust the meaning of fields and reporting formats. Whenever the rules change, traditional reporting systems often require technical personnel to manually compare the differences between the old and new rules, manually modify formulas, fields, or layouts in the templates, and reconfigure the data mapping relationships of multiple source business systems (such as HIS, LIS, core banking systems, and risk control platforms). This process is not only slow and costly in terms of manpower, but it is also prone to misunderstandings or operational errors that can cause templates to become disconnected from the rules, leading to data inconsistencies, logical errors, or formatting inconsistencies, which in severe cases may result in regulatory penalties or biased clinical decisions.

[0004] Therefore, there is an urgent need for a technical solution that can dynamically evolve report templates based on changes in reporting rules and intelligently generate target reports based on multi-source heterogeneous data, so as to achieve cross-domain, highly reliable, and fully automated report generation capabilities. Summary of the Invention

[0005] In view of the above, it is necessary to provide a method for automatic report generation. The purpose of this method is to provide a method that can dynamically evolve report templates according to changes in reporting rules and intelligently generate target reports based on multi-source heterogeneous data.

[0006] Firstly, a method for automatically generating reports is provided, including: When the preset report submission period is reached, raw business data is obtained from the preset business system, and the raw business data is standardized to obtain standard data. Obtain the most recently successfully sent historical report and the sending timestamp of the historical report, as well as the latest reporting rule in the reporting rule base and the publication timestamp of the latest reporting rule. The reporting rule base stores multiple reporting rules in chronological order of publication timestamps. When the publication timestamp is later than the sending timestamp of the historical report, obtain the historical reporting rule whose publication timestamp is closest to the latest reporting rule, extract the filling fields, calculation logic, verification conditions and format requirements of the latest reporting rule and the historical reporting rule respectively, and use the extracted content as rule elements; By comparing the rule elements corresponding to the latest reporting rule with those of the historical reporting rule, at least one rule element that has been changed, added, or deleted is identified, and a template modification operation is generated based on the identified rule element changes to guide the modification of the report template. Based on the template modification operation, a template modification instruction is generated. The template modification instruction is then used to automatically modify the field cells in the report template used by the historical report that correspond to the template modification operation, thereby generating the target template. Based on the latest reporting rules and the target template, report data that meets the requirements of the latest reporting rules is extracted from the standard data, and the report data is filled into the corresponding position in the target template to generate the target report.

[0007] Secondly, an automatic report generation device is provided, comprising: The processing module is used to obtain raw business data from a preset business system when the preset type of report submission period is reached, and to perform standardization processing on the raw business data to obtain standard data. The acquisition module is used to acquire the most recently successfully sent historical report and the sending timestamp of the historical report, as well as the latest reporting rule in the reporting rule base and the publication timestamp of the latest reporting rule. The reporting rule base stores multiple reporting rules stored in chronological order of publication timestamps. The extraction module is used to obtain the historical reporting rule whose publication timestamp is closest to the latest reporting rule when the publication timestamp is later than the sending timestamp of the historical report, and extract the filling fields, calculation logic, verification conditions and format requirements of the latest reporting rule and the historical reporting rule respectively, and use the extracted content as rule elements; The comparison module is used to compare the rule elements corresponding to the latest reported rule with the historical reported rule, identify at least one rule element that has been changed, added or deleted, and generate template modification operations to guide the modification of the report template based on the identified rule element changes. The modification module is used to generate a template modification instruction based on the template modification operation, and to automatically modify the field cells in the report template used by the historical report that correspond to the template modification operation using the template modification instruction, so as to generate a target template. The filling module is used to extract reporting data that meets the requirements of the latest reporting rules from the standard data according to the latest reporting rules and the target template, and fill the reporting data into the corresponding position in the target template to generate a target report.

[0008] Thirdly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described automatic report generation method.

[0009] Fourthly, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the above-described automatic report generation method.

[0010] Compared to existing technologies, this invention accurately determines whether an updated reporting rule has taken effect by comparing the timestamp of the latest rule release with the timestamps of historical report submissions. This timestamp-driven decision-making logic avoids blindly executing template modification processes, triggering subsequent operations only when the rule has indeed been updated, significantly improving system response efficiency and reducing unnecessary computational overhead.

[0011] By comparing the differences between the latest reported rules and the corresponding rule elements of the historical reported rules, template modification operations are obtained. This achieves intelligent conversion from natural language rules to executable change instructions, automatically identifying the addition, deletion, or logical changes of fields and generating structured template modification operations. Based on the template modification instructions, precise addition, modification, or deletion operations are performed on field units in historical templates, automatically generating target templates that conform to the latest rules. The entire process requires no manual intervention, preserving the professional layout and typesetting of the original template while ensuring complete structural consistency with the new rules.

[0012] The latest rules' calculation logic and validation conditions generate accurate reporting data from standard data, and precisely match the positions in the target template using field identifiers for filling, while retaining the original professional styles such as fonts, borders, and conditional formatting. The final output target report meets the reporting requirements in both data logic and visual presentation.

[0013] This invention can dynamically evolve report templates according to changes in reporting rules and intelligently generate target reports based on multi-source heterogeneous data, enabling it to meet the high-frequency and complex reporting needs in fields such as finance and healthcare with high reliability and efficiency. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of an application environment for an automatic report generation method according to an embodiment of the present invention; Figure 2This is a flowchart illustrating an embodiment of the automatic report generation method provided by the present invention. Figure 3 This is a schematic diagram of a module of an automatic report generation device provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of a computer device according to an embodiment of the present invention; Figure 5 This is another structural schematic diagram of a computer device according to one embodiment of the present invention.

[0015] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort are within the scope of protection of this invention.

[0017] It should be noted that the descriptions involving "first," "second," etc., in this invention are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0018] The automatic report generation method provided in this embodiment of the invention can be applied to, for example... Figure 1 In this application environment, the client communicates with the server via a network. When the preset report submission period arrives, the server retrieves raw business data from a preset business system, standardizes the raw business data to obtain standard data, and retrieves the most recently successfully sent historical report and its sending timestamp, as well as the latest reporting rule in the reporting rule base and its publication timestamp. The reporting rule base stores multiple reporting rules in chronological order of their publication timestamps. When the publication timestamp is later than the sending timestamp of the historical report, the historical reporting rule whose publication timestamp is closest to the latest reporting rule is obtained. The fields, calculation logic, verification conditions, and format requirements of the latest reporting rule and the historical reporting rule are extracted respectively, and the extracted content is used as rule elements. The rule elements corresponding to the latest reporting rule and the historical reporting rule are compared to identify at least one rule element that has changed, been added, or been deleted. Based on the identified rule element changes, a template modification operation is generated to guide the modification of the report template. A template modification instruction is generated according to the template modification operation. The template modification instruction is used to automatically modify the field cells in the report template used by the historical report that correspond to the template modification operation to generate a target template. According to the latest reporting rule and the target template, the reporting data that meets the requirements of the latest reporting rule is extracted from the standard data, and the reporting data is filled into the corresponding position in the target template to generate the target report.

[0019] This invention targets fields such as finance and healthcare. By comparing the timestamp of the latest rule release with the timestamps of historical report submissions, it accurately determines whether an updated reporting rule has taken effect. This timestamp-driven decision-making logic avoids blindly executing template modification processes, triggering subsequent operations only when the rule has indeed been updated, significantly improving system response efficiency and reducing unnecessary computational overhead.

[0020] By analyzing the differences between the latest and historical reporting rules, template modification operations are derived. This achieves intelligent conversion from natural language rules to executable change instructions, automatically identifying the addition, deletion, or logical changes of fields and generating structured template modification operations. Based on these template modification instructions, precise addition, modification, or deletion operations are performed on field units in historical templates, automatically generating target templates that conform to the latest rules. The entire process requires no manual intervention, preserving the professional layout and formatting of the original template while ensuring complete structural consistency with the new regulations.

[0021] The latest rules' calculation logic and validation conditions generate accurate reporting data from standard data, and precisely match the positions in the target template using field identifiers for filling, while retaining the original professional styles such as fonts, borders, and conditional formatting. The final output target report meets the reporting requirements in both data logic and visual presentation.

[0022] Reference Figure 2 The diagram shown is a flowchart illustrating an automatic report generation method according to an embodiment of the present invention. This method is executed by a device.

[0023] In this embodiment, a method for automatically generating reports includes: S1. When the report submission period of the preset type is reached, the original business data is obtained from the preset business system, and the original business data is standardized to obtain standard data. In this embodiment, when the current time is detected to be within the reporting cycle of a certain preset type of report (such as monthly business analysis report, quarterly capital adequacy report, annual medical insurance settlement summary report, etc.), the data collection process is automatically triggered.

[0024] Based on the data source configuration associated with this report type, raw business data can be retrieved from one or more preset business systems. For example, in a fintech scenario, account balance data can be obtained from the core banking system, risk-weighted asset (RWA) data from the risk management system, and net profit data from the financial system; or, in a healthcare scenario, patient records can be obtained from the hospital information system (HIS), test results from the laboratory information system (LIS), and expense details from the medical insurance settlement platform.

[0025] Because each business system is built independently, their definitions, units of measurement, statistical standards, or data formats for the same business concept often differ (for example, the financial system records profits in "yuan," while the risk control system stores capital data in "ten thousand yuan"; the HIS uses ICD-10 encoding for "discharge diagnosis," while the regional platform requires the use of local extended codes). Directly using such heterogeneous data to generate reports will lead to inconsistent standards, calculation errors, or verification failures.

[0026] In one embodiment, the standardization process of the original business data to obtain standard data includes: Identify the business system and field meaning corresponding to each data item in the original business data; Based on the preset cross-system indicator mapping relationship, data items from different business systems that correspond to the same reporting indicator but have different field meanings are converted into standard data that conforms to the unified reporting standard.

[0027] In one embodiment, the preset cross-system indicator mapping relationship is constructed in the following way: Obtain the meaning of fields and corresponding reporting rules for data items in each business system; Based on the preset reporting indicator ontology model, the meaning of the field is semantically matched with the reporting rules to determine the unified reporting indicator to which each data item belongs. The correspondence between the business system, the data item, the meaning of the field, and the unified reporting indicator is stored as the cross-system indicator mapping relationship.

[0028] This invention constructs a cross-system indicator mapping relationship, which is a key bridge to bridge the semantic gap between "multi-source heterogeneous business data" and "unified regulatory reporting standards". It enables the system to automatically understand, align and transform data, thereby achieving efficient, accurate and scalable intelligent reporting.

[0029] Each original data item is accompanied by a metadata tag during transmission, or its context information is associated through the data source configuration table. The meaning of each data item's fields is clarified based on the context information, such as "whose data is it, what it represents, how it is calculated, and in what unit / format it is expressed", thus completing the identification of "field meaning".

[0030] The system invokes a pre-built cross-system indicator mapping table, which is stored in a structured format. It iterates through all raw data items, matching them based on their "source system + source field" or semantic similarity to find the corresponding reporting indicator ID. For multiple data items matching the same reporting indicator ID but with different source definitions, the following standardization operations are performed: Unit unification: convert "yuan" to "ten thousand yuan" (divided by 10,000); Scope alignment: if subsidiary data includes minority shareholder profit and loss, but the standard scope requires its exclusion, then the system uses a pre-defined allocation rule for splitting; Encoding mapping: map the ICD-10 code "I10" to the regional reporting standard code "HTN_UNSP"; Missing information completion: if a system does not provide a "complication identifier," but the reporting scope requires it, then the system infers and fills in the default value based on the patient's medical record text using an NLP model. Finally, all raw data items are converted into standard data with consistent structure, semantic alignment, and conformity to the unified reporting scope. By converting raw business data from different business systems into standardized data conforming to unified reporting standards, this invention effectively solves the inconsistencies in field meanings, units of measurement, coding systems, and format specifications of multi-source heterogeneous data. It ensures consistent reporting standards, avoids calculation errors caused by semantic ambiguity, and supports automatic cross-system data fusion, aggregating multi-source information without manual intervention.

[0031] S2. Obtain the most recently successfully sent historical report and the sending timestamp of the historical report, as well as the latest reporting rule in the reporting rule base and the publication timestamp of the latest reporting rule. The reporting rule base stores multiple reporting rules stored in chronological order of publication timestamps. In this embodiment, a pre-defined report sending record library is accessed. This library stores metadata about each report submission in a structured format (such as a relational database table or log index), including: report type, report file identifier, sending status, recipient, and sending timestamp. In a healthcare scenario, such reports may include infectious disease reports, medical insurance settlement statements, and electronic medical record quality assessment data submitted periodically by hospitals to health authorities. In a fintech scenario, these reports may cover compliance data such as anti-money laundering (AML) transaction reports, capital adequacy reports, and customer risk rating summaries submitted by banks or payment institutions to regulatory agencies. Records are filtered by report type, and the record with a "successful" sending status and the largest sending timestamp is retrieved as the "most recently successfully sent historical report," and its corresponding sending timestamp is extracted.

[0032] Access the reporting rule base, which stores multiple reporting rules in chronological order of their publication timestamps. It manages all types of reporting rule versions. Each rule record includes: rule ID, applicable report type, rule content (such as field definitions and calculation logic in JSON / YAML format), publication timestamp, and publishing organization. Based on the type of the report to be generated, filter out all matching rule versions, select the one with the latest publication timestamp as the "latest reporting rule," and retrieve its publication timestamp.

[0033] This invention can automatically and accurately identify whether rules have changed by obtaining the most recently successfully sent historical report and its sending timestamp, as well as the latest reported rules in the reporting rule base and their publication timestamp. This allows the system to decide whether to initiate the difference analysis and template evolution process, avoid unnecessary duplication of processing, and improve system response efficiency and resource utilization.

[0034] S3. When the publication timestamp is later than the sending timestamp of the historical report, obtain the historical reporting rule whose publication timestamp is closest to the latest reporting rule, extract the filling fields, calculation logic, verification conditions and format requirements of the latest reporting rule and the historical reporting rule respectively, and use the extracted content as rule elements; In this embodiment, when the publication timestamp of the latest reporting rule is later than the sending timestamp of the historical report, it is considered that the reporting rule has changed, and further analysis of the changes is needed to guide the template update. All historical rule versions matching the current report type are retrieved from the reporting rule library and sorted in descending order of publication timestamp. The version with the publication timestamp closest to the latest reporting rule is selected as the historical reporting rule.

[0035] When the timestamp of the latest reporting rule is later than the timestamp of the historical report, it indicates that the reporting rule has been updated and further analysis of the changes is needed to guide the template evolution.

[0036] The system retrieves all historical rule versions matching the current report type from the reporting rule base and sorts them in reverse chronological order by publication timestamp. The historical version with the publication timestamp closest to the latest reporting rule is selected as the historical reporting rule. Structured parsing is then performed on both the latest and historical reporting rules. This structured parsing process is based on a pre-defined rule pattern library (containing domain keywords, regular expression templates, and semantic parsing models). It identifies and extracts core content such as data entry fields, calculation logic, validation conditions, and format requirements from the rule text of both the latest and historical reporting rules. This core content is then uniformly categorized into rule elements and stored in a structured form: Data entry fields: such as "capital adequacy ratio" and "primary diagnostic code"; Calculation logic: such as "qualified capital / risk-weighted assets" and "drug costs + examination fees"; Validation conditions: such as "≥8%" and "length 3–7 characters"; Format requirements: such as "retain two decimal places" and "enter in row 5 of Appendix II".

[0037] This invention models four core dimensions—report fields, calculation logic, validation conditions, and format requirements—as rule elements. This allows the system to establish a fine-grained semantic alignment mechanism across multiple rule versions, effectively handling complex change scenarios such as synonym substitutions (e.g., "must not be lower than" versus "≥"), structural adjustments (e.g., indicator splitting or merging), and caliber updates (e.g., adjustments to risk weight calculation methods). This not only significantly improves the accuracy of difference identification but also ensures that subsequent template modifications and data generation strictly adhere to the latest compliance requirements.

[0038] S4. Compare the rule elements corresponding to the latest reporting rule with the historical reporting rule, identify at least one rule element that has been changed, added or deleted, and generate a template modification operation to guide the modification of the report template based on the identified rule element changes. In one embodiment, comparing the rule elements corresponding to the latest reported rule with the historically reported rule to identify at least one rule element that has been changed, added, or deleted includes: The semantic similarity of the rule elements of the latest reporting rule and the historical reporting rule is calculated. When the semantic similarity between two rule elements from the latest reporting rule and the historical reporting rule is higher than a preset threshold, they are determined to be the same rule element, and are marked as changed when the content is inconsistent. When a rule element in the latest reported rule does not have a corresponding rule element in the historically reported rules with a semantic similarity higher than the preset threshold, it is marked as newly added; When a rule element in the historically reported rule does not have a corresponding rule element in the latest reported rule with a semantic similarity higher than the preset threshold, it is marked as deleted.

[0039] A semantic comparison is performed between the rule elements of the latest reported rules and those of historically reported rules. Specifically, a semantic encoding model (such as a fine-tuned Sentence-BERT) is used to convert each rule element into a vector, and the cosine similarity between the new and old rule elements is calculated. A preset threshold (e.g., 0.85) is set. If a historical element and a new element have a similarity score higher than the threshold, they are considered the same rule element. If their content differs (e.g., formula or threshold change), they are marked as changed. If an element in the new rule has no matching item with a similarity score higher than the threshold in the historical rules, it is marked as added. If an element in the historical rules has no matching item in the new rule, it is marked as deleted. For example, in a healthcare scenario: The new rule adds the element {"type": "reporting field", "content":"carbon emission intensity"}, and there is no related field in the historical rules, with a maximum similarity score of only 0.45 (<0.85) → marked as added. A "paper medical record submission flag" exists in the historical rules, but it has been removed in the new rules, and there is no new element matching it → marked as deleted.

[0040] This invention, by identifying at least one rule element that has been changed, added, or deleted, serves as a crucial link between rule semantic understanding and automated template execution. It enables the system to possess human-level rule interpretation capabilities while executing changes with machine-level precision, thus truly achieving an intelligent reporting loop where "once a rule is published, the report automatically adapts." Without this precise identification mechanism, subsequent template modifications would lack a basis, and the entire automated process would degenerate into rudimentary redoing or reliance on manual intervention.

[0041] In one embodiment, based on the identified change type of rule element, a corresponding template modification operation is generated, including: If it is a newly added rule element, a new class template modification operation is generated, which includes the target field identifier, field name, data type, calculation logic, and insertion position. If it is a rule element to be modified, a modification class template modification operation containing the target field identifier and the updated rule expression will be generated; If it is a rule element to be deleted, a deletion class template modification operation containing the target field identifier is generated.

[0042] Each template modification operation is used to locate the field cells in the report template and perform the corresponding modification operations.

[0043] This invention enables the automatic transformation from unstructured regulatory rules to structured, executable template modification operations, ensuring that template evolution accurately reflects rule changes—new fields can be inserted based on semantic positioning, modified content strictly aligns with the latest logic, and deleted items thoroughly eliminate redundant structures. Thus, the system guarantees dynamic consistency of report templates in terms of fields, formulas, formats, and regulatory requirements without manual intervention, providing key technical support for intelligent, reliable, and unattended reporting in highly compliant scenarios such as finance and healthcare.

[0044] In one embodiment, the step of extracting the data entry fields, calculation logic, verification conditions, and format requirements of the latest reporting rule and the historical reporting rule, and using the extracted content as rule elements, includes: Obtain the rule texts of the latest reporting rule and the historical reporting rule; Identify semantic units in the rule text, including indicator names, numerical expressions, conditional statements, and formatting instructions; According to the preset rule pattern library, the semantic units are classified and mapped to corresponding rule element types, including: mapping the indicator name to the filling field, mapping the numerical expression to the calculation logic, mapping the condition judgment statement to the validation condition, and mapping the typesetting instruction to the format requirements.

[0045] A hybrid parsing strategy (combining a rule engine and a lightweight NLP model) is used to scan rule text and identify four types of key semantic units, including indicator names, numerical expressions, conditional statements, and typesetting instructions.

[0046] Indicator Name: A noun phrase representing a business concept, such as "capital adequacy ratio"; Numerical Expression: A formula containing mathematical operations or functions, such as "qualified capital / risk-weighted assets"; Conditional Statement: Constraints containing comparisons or logical relationships, such as "must not be lower than 8%" or "if age ≥ 65 years, then mark as an elderly patient"; Layout Instructions: Descriptions indicating the position or format of a field in the report, such as "enter row 5 of Table 3" or "retain two decimal places".

[0047] The default rule pattern library is a structured configuration table that defines the mapping relationship between semantic unit types and rule element types. Specifically, it maps "indicator name" to data entry fields, "numerical expression" to calculation logic, "conditional judgment statement" to validation conditions, and "formatting instructions" to format requirements.

[0048] This invention parses unstructured rule text into four standardized rule elements (fill-in fields, calculation logic, validation conditions, and format requirements), thereby unifying the rule expression and enabling machines to consistently understand rules from different sources and with different wording. It supports fine-grained comparison, accurately identifying which part has changed (whether the formula has changed, the threshold has been adjusted, or a new field has been added).

[0049] In one embodiment, the method further includes: When the publication timestamp is not later than the sending timestamp, obtain the report template used by the historical report; Based on the latest reporting rules and the report template, extract reporting data that meets the requirements of the latest reporting rules from the standard data; The reported data is filled into the corresponding positions in the report template to generate the target report.

[0050] If the publication timestamp of the latest reporting rule is not later than the sending timestamp of the historical report, it indicates that the rule has not been updated, and the report template used in the previous successful submission is directly reused. This means the report template has been verified as compliant, and its structure, fields, and format are all valid. The system still uses the latest reporting rule as the basis, performs structured parsing, extracts the calculation logic and verification conditions, and combines standard data to generate reporting data that conforms to the current rule, ensuring the accuracy of the data processing logic. Then, the reporting data is filled into the corresponding positions in the template, and the target report is output for verification and submission. This invention directly reuses historical compliant templates when the rule has not been updated, while still processing data and filling it into the template according to the latest reporting rule, thereby avoiding unnecessary template modifications, improving generation efficiency, and ensuring that the report content is consistent with the current rule.

[0051] This invention accurately identifies rule changes, additions, or deletions by analyzing the differences between the latest and historical reporting rules, overcoming the misjudgment problems of traditional literal comparison under synonym substitution and expression differences; it automatically generates structured template modification operations, drives intelligent template evolution, supports field addition and deletion, logic adjustment and verification updates, significantly improves the reporting system's responsiveness to rule changes and compliance reliability, and is a key support for realizing intelligent, unattended report generation.

[0052] S5. Generate a template modification instruction based on the template modification operation, and use the template modification instruction to automatically modify the field cells in the report template used by the historical report that correspond to the template modification operation, and generate the target template. In one embodiment, the template modification instructions include add instructions, modify instructions, and delete instructions, and generating template modification instructions based on the template modification operation includes: Parse the operation type and target field identifier for each template modification operation; When the operation type is "add", an "add" instruction containing the target field identifier and the insertion position is generated; When the operation type is modification, a modification instruction containing the target field identifier and the updated rule expression is generated; When the operation type is deletion, a deletion instruction containing the target field identifier is generated; After obtaining one or more template modification operations, read the metadata of each template modification operation, extract the operation type (operation_type) and target field identifier (field_id). The target field identifier is used to uniquely associate field units in the template.

[0053] (1) When the operation type is "Add", an add instruction is generated. The add instruction includes field definition, insertion position and default value. The field definition consists of field name, data type and calculation logic; the insertion position is obtained from the typesetting instructions extracted in the rule parsing stage; default value: if the rule does not provide actual data, it can be set to 0, empty string or preset placeholder.

[0054] (2) When the operation type is "Modify", a modification instruction is generated. The modification instruction includes the target field identifier and the updated rule expression. Target field identifier: Specifies the field to be updated; Updated rule expression: The new calculation logic or validation condition from the difference analysis results, used to replace the original formula in the template.

[0055] (3) When the operation type is "delete", a deletion instruction is generated. The deletion instruction only needs to contain the target field identifier, indicating that the target field and its associated display elements should be removed.

[0056] This invention stores all generated template modification instructions in a structured format (such as JSON or internal objects) and passes them to the template modification module. These instructions precisely describe the atomic-level operations on the report template, ensuring that the template can automatically and unambiguously evolve into the target template that conforms to the latest rules.

[0057] In one embodiment, the step of automatically modifying the field cells corresponding to the template modification operation in the report template used by the historical report using the template modification instruction to generate the target template includes: Identify each field cell in the report template, and associate each field cell with a field identifier; Based on the target field identifier in the template modification instruction, match the field cells in the report template that have the same field identifier; When the template modification instruction is a new instruction, a new field cell is created at the insertion position, and its field name, data type and calculation logic are configured. When the template modification instruction is a modification instruction, the calculation logic or verification rule of the field unit is updated to the updated rule expression; When the template modification instruction is a delete instruction, the field unit and its associated display elements are removed; Save the modified report template as the target template.

[0058] The report template file associated with the successfully loaded report was loaded. The report template file is stored in a structured format (such as Excel with named ranges, Word with content controls, JSON form definition, or XML Schema), which contains multiple field cells. Each field cell is associated with a unique field identifier (such as CAPITAL_ADEQUACY, PRIMARY_DIAG_CODE) for programmatic location and operation.

[0059] Process each template modification command sequentially, and perform the following actions based on the operation type of the template modification command: (1) When a new instruction is received, the insertion position is parsed, a new field cell is created at the specified position in the template, and the field name, data type, calculation logic and display style are configured (inheriting the default template format or setting according to the rules).

[0060] (2) When a modification instruction is received, the corresponding field unit in the template is located according to the target field identifier, and its calculation logic or verification rule is replaced with the updated rule expression provided in the instruction.

[0061] (3) When a deletion instruction is received, locate the field cell according to the target field identifier and remove the cell and its associated display elements, including cell content, formulas, data validation, conditional formatting, row / column headers (if only serving the field) and bound named ranges or bookmarks.

[0062] After executing all template modification instructions, this invention saves the modified template as a new version, named "Target Template," and records its association with the latest reporting rules. This target template retains the layout, style, and unchanged fields of the original template, making precise adjustments only to the differing parts to ensure format compliance and structural integrity.

[0063] Once reporting rules are updated (such as adjustments to financial regulatory indicators or changes to medical insurance codes), traditional templates, if not updated accordingly, will result in inconsistent reporting formats, missing fields, or logical errors. The target template of this invention, however, automatically maps rule changes to ensure that field definitions, calculation formulas, and verification conditions are completely consistent with the latest requirements, fundamentally avoiding compliance risks.

[0064] Traditional methods require technicians to manually compare rules and modify Excel / Word templates, which is time-consuming and prone to errors (such as correcting incorrect cells or omitting formulas). In contrast, the target template of this invention is generated automatically and programmatically by the system, requiring no manual intervention, significantly reducing maintenance costs and eliminating human error.

[0065] S6. Based on the latest reporting rules and the target template, extract reporting data that meets the requirements of the latest reporting rules from the standard data, and fill the reporting data into the corresponding position in the target template to generate a target report.

[0066] In one embodiment, extracting reporting data that conforms to the requirements of the latest reporting rules from the standard data according to the latest reporting rules and the target template includes: Extract the calculation logic and verification conditions from the latest reporting rules, and generate data processing instructions based on the calculation logic and verification conditions; Based on the field identifier associated with each field unit in the target template, the corresponding data item is matched from the standard data; The data processing instructions are used to perform calculation, transformation, or aggregation operations on the matched data items to generate reporting data that is consistent with the field structure of the target template.

[0067] The text content of the latest reported rules is structured and parsed to extract data processing-related elements (calculation logic and verification conditions). Calculation logic includes, for example, "Capital Adequacy Ratio = Qualified Capital / Risk-Weighted Assets"; verification conditions include, for example, "Liquidity Coverage Ratio ≥ 100%" and "Age ≥ 0 and ≤ 150". These elements are then converted into instructions that can be executed by the data engine, such as parsing formulas into expression trees or script functions, and converting verification conditions into assertion rules.

[0068] The system iterates through all field cells in the target template, obtaining a unique field identifier associated with each cell, and searches for data items with the same field identifier in the standard dataset. It establishes a mapping relationship between template fields and standard data values ​​based on the field identifiers, applies executable instructions to the matched data items, and performs calculations, transformations, or aggregations. Examples of calculations include arithmetic or logical operations on multiple fields; transformations include unit conversions (yuan → ten thousand yuan) and formatting (keeping two decimal places); and aggregations include summarizing by institution and grouping by disease. Finally, it generates reporting data with a field structure identical to the target template, with each field containing a field identifier and calculation result.

[0069] This invention transforms the calculation logic and verification conditions in the rules into executable data processing instructions, which are then applied to the matched standard data items. The system can automatically complete the required operations (such as summation and ratio), format conversion (such as unit unification and decimal truncation), or data aggregation (such as summarization by organization) for the fields, thereby outputting structured reporting data that corresponds one-to-one with the template fields and meets the latest regulatory requirements, ensuring the accuracy, compliance, and automated generation capabilities of the reports.

[0070] In one embodiment, the step of filling the reported data into the corresponding position in the target template to generate the target report includes: Iterate through each field unit in the target template and obtain the field identifier associated with each field unit; Based on the field identifier, retrieve the corresponding data content from the reported data; Write the retrieved data content into the preset display position of the field unit in the target template; The original formatting attributes of the target template are retained, including font, border, cell style, and conditional formatting rules; The completed target template is output as the target report.

[0071] Load the generated target template (such as an Excel file, Word document, or JSON form definition) and iterate through all the fillable field cells. Each field cell has been bound with a unique field identifier (such as CAPITAL_ADEQUACY, PRIMARY_DIAG_CODE) during the template creation phase, usually in the form of an Excel named range, a Word bookmark, or a JSON key.

[0072] From the generated set of reported data (structured key-value pairs), the corresponding data content (i.e., numerical value, text, or calculation result) is found based on the field identifier. The found data content is written into the preset display position of the field cell in the template, while strictly preserving the original format attributes. After filling, the modified document is fixed into an uneditable submission format (such as PDF, encrypted Excel, or XML) and named the target report for subsequent verification, approval, or submission. This invention accurately and automatically fills the structured reported data into the corresponding positions while maintaining the original format and style of the target template, ensuring that the generated report is both accurate in content and compliant in format, and can be directly used for submission without manual intervention or post-processing adjustments.

[0073] In one embodiment, after the target report, the method further includes: Perform multi-dimensional intelligent verification on the target report; When the target report passes the verification, the target report is sent to the preset user; When the target report fails the verification, analyze the reasons for the verification failure and generate a repair report containing data correction suggestions or calculation logic adjustment suggestions; Based on the repair report, the standard data or reported data generation logic is corrected and the target report is regenerated until the target report passes the verification.

[0074] After the target report is generated, multi-dimensional intelligent verification is automatically performed to ensure that its content is accurate, logically compliant, and formatted correctly. If the verification fails, an intelligent diagnosis and repair mechanism is activated, forming a closed loop, until a fully compliant target report is generated.

[0075] The validation engine performs multi-dimensional checks in parallel based on a pre-defined validation rule library (which can be automatically extracted from reported rules) to generate structured validation results. These multi-dimensional checks include, but are not limited to: logical consistency validation (checking for errors in the interrelationships between fields), rule compliance validation (verifying based on the validation conditions in the latest reported rules), data integrity validation (checking whether required fields are empty or missing), and format standardization validation (verifying whether numerical precision, coding standards, units, etc., meet the requirements).

[0076] If all verification items pass, the target report will be encrypted, packaged, and automatically sent to the preset user (such as the finance manager) through a preset channel (such as email, regulatory reporting platform API, or internal approval system).

[0077] If any verification item fails, the system automatically analyzes the cause of the failure and generates a structured repair report, including: failure location, failure type, root cause prediction, and correction suggestions. Based on the repair report, the system automatically triggers the correction process: if the problem originates from standard data, it backtracks to stage S1 to adjust cross-system mapping relationships or data cleaning rules; if the problem originates from the reported data generation logic, it updates data processing instructions (such as correcting formulas); after correction, it re-executes the data filling and report generation in stage S5 and performs verification again. This process can be configured with a maximum number of retries (e.g., 3 times) to avoid infinite loops. Once the target report passes all verifications, the loop terminates and the sending process begins.

[0078] This invention constructs a closed-loop intelligent verification and self-repair mechanism to ensure that the final submitted reports fully comply with regulatory requirements in terms of content, logic, and format, significantly reducing the cost of manual review and the risk of submission failure. By promptly identifying errors through multi-dimensional verification and automatically or semi-automatically correcting data or calculation logic based on repair suggestions, the system not only improves the accuracy of reports but also continuously optimizes the data processing flow, enhancing the overall reliability and compliance efficiency of submission.

[0079] like Figure 3 The diagram shown is a schematic diagram of a report automatic generation device provided in an embodiment of the present invention.

[0080] The automatic report generation device 100 of this invention can be installed in a device. Depending on the functions implemented, the automatic report generation device 100 may include a processing module 110, an acquisition module 120, an extraction module 130, a comparison module 140, a modification module 150, and a filling module 160. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the device processor and perform a fixed function, and are stored in the device's memory.

[0081] In this embodiment, the functions of each module / unit are as follows: Processing module 110 is used to obtain raw business data from a preset business system when the preset type of report submission period is reached, and to perform standardization processing on the raw business data to obtain standard data. The acquisition module 120 is used to acquire the most recently successfully sent historical report and the sending timestamp of the historical report, as well as the latest reporting rule in the reporting rule base and the publication timestamp of the latest reporting rule. The reporting rule base stores multiple reporting rules stored in chronological order according to their publication timestamps. The extraction module 130 is used to obtain the historical reporting rule whose publication timestamp is closest to the latest reporting rule when the publication timestamp is later than the sending timestamp of the historical report, extract the filling fields, calculation logic, verification conditions and format requirements of the latest reporting rule and the historical reporting rule respectively, and use the extracted content as rule elements; The comparison module 140 is used to compare the rule elements corresponding to the latest reported rule with the historical reported rule, identify at least one rule element that has been changed, added or deleted, and generate a template modification operation to guide the modification of the report template based on the identified rule element changes. Modification module 150 is used to generate template modification instructions based on the template modification operation, and use the template modification instructions to automatically modify the field cells in the report template used by the historical report that correspond to the template modification operation, so as to generate a target template; The filling module 160 is used to extract reporting data that meets the requirements of the latest reporting rules from the standard data according to the latest reporting rules and the target template, and fill the reporting data into the corresponding position in the target template to generate a target report.

[0082] In one embodiment, the comparison module 140 is specifically used for: The semantic similarity of the rule elements of the latest reporting rule and the historical reporting rule is calculated. When the semantic similarity between two rule elements from the latest reporting rule and the historical reporting rule is higher than a preset threshold, they are determined to be the same rule element, and are marked as changed when the content is inconsistent. When a rule element in the latest reported rule does not have a corresponding rule element in the historically reported rules with a semantic similarity higher than the preset threshold, it is marked as newly added; When a rule element in the historically reported rule does not have a corresponding rule element in the latest reported rule with a semantic similarity higher than the preset threshold, it is marked as deleted.

[0083] In one embodiment, the modification module 150 is specifically used for: Parse the operation type and target field identifier for each template modification operation; When the operation type is "add", an "add" instruction containing the target field identifier and the insertion position is generated; When the operation type is modification, a modification instruction containing the target field identifier and the updated rule expression is generated; When the operation type is deletion, a deletion instruction containing the target field identifier is generated.

[0084] In one embodiment, the modification module 150 is specifically used for: Identify each field cell in the report template, and associate each field cell with a field identifier; Based on the target field identifier in the template modification instruction, match the field cells in the report template that have the same field identifier; When the template modification instruction is a new instruction, a new field cell is created at the insertion position, and its field name, data type and calculation logic are configured. When the template modification instruction is a modification instruction, the calculation logic or verification rule of the field unit is updated to the updated rule expression; When the template modification instruction is a delete instruction, the field unit and its associated display elements are removed; Save the modified report template as the target template.

[0085] In one embodiment, the filling module 160 is specifically used for: Extract the calculation logic and verification conditions from the latest reporting rules, and generate data processing instructions based on the calculation logic and verification conditions; Based on the field identifier associated with each field unit in the target template, the corresponding data item is matched from the standard data; The data processing instructions are used to perform calculation, transformation, or aggregation operations on the matched data items to generate reporting data that is consistent with the field structure of the target template.

[0086] In one embodiment, the processing module 110 is specifically used for: Identify the business system and field meaning corresponding to each data item in the original business data; Based on the preset cross-system indicator mapping relationship, data items from different business systems that correspond to the same reporting indicator but have different field meanings are converted into standard data that conforms to the unified reporting standard.

[0087] In one embodiment, the processing module 110 is specifically used for: Obtain the meaning of fields and corresponding reporting rules for data items in each business system; Semantic matching is performed between the meaning of the fields and the reporting rules to determine the unified reporting indicator to which each data item belongs; The correspondence between the business system, the data item, the meaning of the field, and the unified reporting indicator is stored as the cross-system indicator mapping relationship.

[0088] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 4 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with external clients via a network connection. When the computer program is executed by the processor, it implements the functions or steps of a server-side automatic report generation method.

[0089] In one embodiment, a computer device is provided, which may be a client, and its internal structure diagram may be as follows: Figure 5As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with an external server via a network connection. When the computer program is executed by the processor, it implements the client-side functions or steps of an automatic report generation method.

[0090] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps: When the preset report submission period is reached, raw business data is obtained from the preset business system, and the raw business data is standardized to obtain standard data. Obtain the most recently successfully sent historical report and the sending timestamp of the historical report, as well as the latest reporting rule in the reporting rule base and the publication timestamp of the latest reporting rule. The reporting rule base stores multiple reporting rules in chronological order of publication timestamps. When the publication timestamp is later than the sending timestamp of the historical report, obtain the historical reporting rule whose publication timestamp is closest to the latest reporting rule, extract the filling fields, calculation logic, verification conditions and format requirements of the latest reporting rule and the historical reporting rule respectively, and use the extracted content as rule elements. Compare the rule elements corresponding to the latest reporting rule and the historical reporting rule to identify at least one rule element that has changed, been added or deleted. Based on the identified rule element changes, generate template modification operations to guide the modification of the report template. Based on the template modification operation, a template modification instruction is generated. The template modification instruction is then used to automatically modify the field cells in the report template used by the historical report that correspond to the template modification operation, thereby generating the target template. Based on the latest reporting rules and the target template, report data that meets the requirements of the latest reporting rules is extracted from the standard data, and the report data is filled into the corresponding position in the target template to generate the target report.

[0091] It should be noted that the functions or steps that can be implemented by the computer-readable storage medium or computer device described above can be referred to the relevant descriptions on the server side and client side in the foregoing method embodiments. To avoid repetition, they will not be described one by one here.

[0092] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0093] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0094] It should be noted that any AI models, software tools, or components not belonging to this company appearing in the embodiments of this application are merely illustrative examples and do not represent actual use. All user personal information involved in the embodiments of this application has been authorized (with the knowledge and consent) by the relevant parties or has been fully authorized by all parties, and the executing entity may obtain it through various legal and compliant means. The collection, storage, use, processing, transmission, provision, and disclosure of the information, data, and signals involved all comply with relevant laws and regulations and do not violate public order and good morals.

[0095] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for automatically generating reports, characterized in that, The method includes: When the preset report submission period is reached, raw business data is obtained from the preset business system, and the raw business data is standardized to obtain standard data. Obtain the most recently successfully sent historical report and the sending timestamp of the historical report, as well as the latest reporting rule in the reporting rule base and the publication timestamp of the latest reporting rule. The reporting rule base stores multiple reporting rules in chronological order of publication timestamps. When the publication timestamp is later than the sending timestamp of the historical report, obtain the historical reporting rule whose publication timestamp is closest to the latest reporting rule, extract the filling fields, calculation logic, verification conditions and format requirements of the latest reporting rule and the historical reporting rule respectively, and use the extracted content as rule elements; By comparing the rule elements corresponding to the latest reporting rule with those of the historical reporting rule, at least one rule element that has been changed, added, or deleted is identified, and a template modification operation is generated based on the identified rule element changes to guide the modification of the report template. Based on the template modification operation, a template modification instruction is generated. The template modification instruction is then used to automatically modify the field cells in the report template used by the historical report that correspond to the template modification operation, thereby generating the target template. Based on the latest reporting rules and the target template, report data that meets the requirements of the latest reporting rules is extracted from the standard data, and the report data is filled into the corresponding position in the target template to generate the target report.

2. The automatic report generation method as described in claim 1, characterized in that, The step of comparing the rule elements corresponding to the latest reported rule with the historical reported rule, and identifying at least one rule element that has been changed, added, or deleted, includes: The semantic similarity of the rule elements of the latest reporting rule and the historical reporting rule is calculated. When the semantic similarity between two rule elements from the latest reported rule and the historical reported rule is higher than a preset threshold, they are determined to be the same rule element, and are marked as changed when the content is inconsistent. When a rule element in the latest reported rule does not have a corresponding rule element in the historically reported rules with a semantic similarity higher than the preset threshold, it is marked as newly added; When a rule element in the historically reported rule does not have a corresponding rule element in the latest reported rule with a semantic similarity higher than the preset threshold, it is marked as deleted.

3. The automatic report generation method as described in claim 1, characterized in that, The template modification instructions include add instructions, modify instructions, and delete instructions. Generating template modification instructions based on the template modification operation includes: Parse the operation type and target field identifier for each template modification operation; When the operation type is "add", an "add" instruction containing the target field identifier and the insertion position is generated; When the operation type is modification, a modification instruction containing the target field identifier and the updated rule expression is generated; When the operation type is deletion, a deletion instruction containing the target field identifier is generated.

4. The automatic report generation method as described in claim 1, characterized in that, The step of automatically modifying the field cells corresponding to the template modification operation in the report template used by the historical report using the template modification command to generate the target template includes: Identify each field cell in the report template, and associate each field cell with a field identifier; Based on the target field identifier in the template modification instruction, match the field cells in the report template that have the same field identifier; When the template modification instruction is a new instruction, a new field cell is created at the insertion position based on the insertion position contained in the template modification instruction, and its field name, data type and calculation logic are configured. When the template modification instruction is a modification instruction, the calculation logic or verification rule of the field unit is updated to the updated rule expression; When the template modification instruction is a delete instruction, the field unit and its associated display elements are removed; Save the modified report template as the target template.

5. The automatic report generation method as described in claim 1, characterized in that, The step of extracting reporting data that meets the requirements of the latest reporting rules from the standard data according to the latest reporting rules and the target template includes: Extract the calculation logic and verification conditions from the latest reporting rules, and generate data processing instructions based on the calculation logic and verification conditions; Based on the field identifier associated with each field unit in the target template, the corresponding data item is matched from the standard data; The data processing instructions are used to perform calculation, transformation, or aggregation operations on the matched data items to generate reporting data that is consistent with the field structure of the target template.

6. The automatic report generation method as described in claim 1, characterized in that, The standardization process for obtaining standard data from the original business data includes: Identify the business system and field meaning corresponding to each data item in the original business data; Based on the preset cross-system indicator mapping relationship, data items from different business systems that correspond to the same reporting indicator but have different field meanings are converted into standard data that conforms to the unified reporting standard.

7. The automatic report generation method as described in claim 6, characterized in that, The preset cross-system indicator mapping relationship is constructed in the following way: Obtain the meaning of fields and corresponding reporting rules for data items in each business system; Semantic matching is performed between the meaning of the fields and the reporting rules to determine the unified reporting indicator to which each data item belongs; The correspondence between the business system, the data item, the meaning of the field, and the unified reporting indicator is stored as the cross-system indicator mapping relationship.

8. An automatic report generation device, characterized in that, The device includes: The processing module is used to obtain raw business data from a preset business system when the preset type of report submission period is reached, and to perform standardization processing on the raw business data to obtain standard data. The acquisition module is used to acquire the most recently successfully sent historical report and the sending timestamp of the historical report, as well as the latest reporting rule in the reporting rule base and the publication timestamp of the latest reporting rule. The reporting rule base stores multiple reporting rules stored in chronological order of publication timestamps. The extraction module is used to obtain the historical reporting rule whose publication timestamp is closest to the latest reporting rule when the publication timestamp is later than the sending timestamp of the historical report, and extract the filling fields, calculation logic, verification conditions and format requirements of the latest reporting rule and the historical reporting rule respectively, and use the extracted content as rule elements; The comparison module is used to compare the rule elements corresponding to the latest reported rule with the historical reported rule, identify at least one rule element that has been changed, added or deleted, and generate template modification operations to guide the modification of the report template based on the identified rule element changes. The modification module is used to generate a template modification instruction based on the template modification operation, and to automatically modify the field cells in the report template used by the historical report that correspond to the template modification operation using the template modification instruction, so as to generate a target template. The filling module is used to extract reporting data that meets the requirements of the latest reporting rules from the standard data according to the latest reporting rules and the target template, and fill the reporting data into the corresponding position in the target template to generate a target report.

9. A device, characterized in that, The device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores an automatic report generation program that can be executed by the at least one processor, the automatic report generation program being executed by the at least one processor to enable the at least one processor to perform the automatic report generation method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores an automatic report generation program, which can be executed by one or more processors to implement the automatic report generation method as described in any one of claims 1 to 7.

Images