A method, system and computer program product for linkage and summary of case handling deadline risks and recording compliance risks in an intranet environment

Abstract

The application discloses a case handling period risk and record compliance risk linkage summary method, system and computer program product under an internal network environment. Unified case identification is used to associate case basic data, period calculation results, record documents, compliance verification reports and operation logs. A period calculation module determines the expiration date, remaining days and period early warning risks from the case filing date, handling period, extension, non-counted period days and handling state. A record compliance verification module uses a hit chain traceable deterministic rule base to obtain record compliance risks from the associated records, and does not rely on the uncertainty output of a large language model. A risk synthesis module maps the above two types of risks with different sources, types and calculation methods into high, medium and low levels, respectively, and takes the higher one to obtain a comprehensive risk level. In response to the events of handling, resuming handling, soft deleting, adding a new record or generating a report, the business semantics are automatically recalculated and refreshed. The whole process is completed on the internal network and the data is not externally transmitted.

Description

Technical Field

[0001] This invention relates to the field of quality management and computer-aided technology for public security law enforcement and case handling, specifically to a method, system, and computer program product that, in an intranet environment physically isolated from the Internet, links and summarizes the risks of case handling time limits and compliance risks of law enforcement records in a deterministic and traceable manner to a unified case dimension, and automatically refreshes with case events. Background Technology

[0002] When handling public security, administrative, and criminal cases, the quality of case handling is simultaneously affected by two completely different types of risks: one is the risk of time limits, such as approaching or exceeding the time limit; the other is the compliance risk of law enforcement record materials, such as missing elements, procedural flaws, inadequate notification of rights and obligations, and incomplete signature and fingerprint confirmation. These two types of risks are highly heterogeneous in terms of data sources, data types, and processing methods—the former originates from structured case registration and extension data, data not included in time limits, and relies on date arithmetic; the latter originates from unstructured natural language text records and relies on rule-based interpretation of the text. Existing information technology methods have the following shortcomings: 1. Case deadlines are usually managed separately by ledgers or case handling systems, while the quality of transcripts depends on manual review or independent document quality inspection. The two are scattered in different systems or manual records, making it difficult to form a comprehensive judgment on the two types of heterogeneous risks in the same case dimension. As a result, the near-expiration deadline and material defects cannot be detected and dealt with in a timely manner. 2. Some automated review schemes for transcripts or documents rely on machine learning models or large language models. Their judgment process is a black box, and the output is uncertain. It is neither interpretable nor can it be reviewed line by line, making it difficult to meet the requirements of traceable judgment of case handling compliance and accountability. Moreover, such schemes mostly rely on cloud services, which are difficult to deploy stably on the public security intranet that is physically isolated from the Internet, has high data security requirements, and does not allow data to be sent out. 3. The calculation of time limits and the quality inspection of records are independent of each other. When the case status changes (such as closed, resumed, soft deletion, batch import, new records, or generation of compliance verification reports), the original risk conclusions cannot be automatically updated. Furthermore, there is a lack of update logic that is bound to the semantics of case handling business. It relies solely on manual triggering or simple status change recalculation, which is prone to statistical errors and omissions.

[0003] 4. Regarding prior art solutions related to this invention, they can be broadly divided into two layers: the first is a single-dimensional management layer for case handling time limits or case handling early warnings, such as the case handling early warning management method and device with publication number CN116384940A, the method for automatically identifying cases not filed within a specified time limit with publication number CN111899132A, and the traffic violation processing method based on deterministic rules with publication number CN111383150B. Such solutions only provide early warnings or verifications around case handling time limits, case filing time limits, or single business rules, without involving record compliance risks, and do not combine time limit risks and record compliance risks in the same case dimension; the second is an automatic review layer for record documents, such as the intelligent record review method based on text semantic understanding with publication number CN112861757A, the case record element extraction method with publication number CN114461779A, and the method for extracting case record elements with publication number CN119271793A. The existing methods for detecting contradictions in written records and the BERT and LSTM-based method for processing written record elements (publication number CN111581975B) generally employ machine learning, semantic models, or large language models. Their judgments are probabilistic, cannot be reviewed item by item, and do not involve linkage with case-handling deadline risks. It is evident that existing technologies either focus solely on the deadline dimension or only on the written record dimension, with the written record dimension primarily employing machine learning, semantic, or large language model approaches. None of these technologies synthesize the two heterogeneous risks—delay warning risk and written record compliance risk—in a deterministic and traceable manner into a comprehensive case risk that updates with case-related business events. For the purposes of this invention, the closest existing technology at the core level is the aforementioned case-handling warning management scheme (CN116384940A), and at the written record level is the aforementioned intelligent written record review scheme (CN112861757A).

[0004] Therefore, there is an urgent need for a technical solution that, in an offline intranet environment, combines the risk of deadline warnings and the risk of record compliance, which are heterogeneous in origin and calculation method, into a comprehensive risk of a unified case dimension in a deterministic and traceable manner, and automatically refreshes with case business events. Summary of the Invention

[0005] (a) Technical problems to be solved This invention provides a method, system, and computer program product for linking and summarizing case handling time limit risks and record compliance risks in an intranet environment, in order to solve the following problems: the dispersion of two types of heterogeneous risks, which cannot be synthesized into a comprehensive judgment at a unified case dimension; the existing automatic review schemes rely on the cloud or uncertainty models, resulting in uninterpretability, difficulty in intranet deployment, and data external transmission; and the risk conclusions cannot be automatically refreshed with case business events and lack refresh logic bound to business semantics.

[0006] (II) Technical Solution The system uses a unified case identifier to link basic case data, time limit calculation results, transcript documents, and their deterministic compliance verification reports and operation logs. The time limit calculation module performs deterministic date calculations on structured case time limit data to obtain time limit warning risks (Category 1 risks), while the transcript compliance verification module performs rule matching on unstructured transcript natural language text using a deterministic compliance rule base to obtain transcript compliance risks (Category 2 risks). The two risks are heterogeneous in origin, data type, and calculation method. The risk synthesis module combines the above two heterogeneous risks into a comprehensive case risk level according to preset synthesis rules. When a case is closed, resumed, softly deleted, a new transcript is added, or a compliance verification report is generated, the comprehensive risk level is automatically recalculated and refreshed according to logic bound to business semantics. The entire process is completed on the local machine on the intranet, which is physically isolated from the Internet, and the compliance determination does not rely on the uncertain output of machine learning models or large language models, and the rule matching chain is traceable. The system is deployed on an intranet physically isolated from the internet and where data is not shared externally. Objectively, it cannot access uninterpretable model services such as large language models deployed in the cloud. This constitutes the technical motivation for this invention to use a local deterministic rule engine instead of machine learning models or large language models for transcript compliance determination. This makes the intranet physical isolation and data non-externalization a technical feature that substantially contributes to the interpretability and traceability of the technical solution, rather than simply a declaration of the usage environment. Specific limitations of the technical solution are as described in the claims.

[0007] (III) Beneficial Effects 1. Unify case dimensions and link two types of heterogeneous risks: Combine the time-limited warning risk and the record compliance risk, which have completely different sources and calculation methods, into the comprehensive risk of the same case, and overcome the omissions caused by only looking at the ledger or only looking at the record; 2. Explainable and auditable: The deadline risk is based on deterministic date calculations, and the compliance judgment of the record is based on a deterministic rule base. Each hit issue can be traced back to the rule category, checkpoint, legal basis and handling suggestions of the rule base entry through its rule identifier, forming a hit chain that can be verified item by item, which cannot be provided by machine learning models or large language models in terms of structure; the risk synthesis adopts two types of heterogeneous risks, which are each normalized to a unified three-level risk level scale, and the higher deterministic weakness is synthesized. The calculation process and the hit basis can be verified item by item, which meets the requirements of case handling compliance and accountability. 3. Automatic linkage refresh and binding of business semantics: Business events such as case completion, recovery, soft deletion, addition of records, and generation of reports trigger automatic recalculation of comprehensive risks. It also handles the removal and inclusion of near-term statistics, the retention of historical reports, and the clearing and restoration of summary records according to business semantics, avoiding manual synchronization omissions and statistical errors. 4. Intranet offline operation and data security: It includes an offline runtime package for deployment, runs locally, and does not transmit data externally, making it compatible with public security intranets that are physically isolated from the Internet; 5. Deterministic verification for the field of record-keeping: Matching rule subsets according to record type, combined with negative context recognition to reduce false alarms, and outputting hierarchical and traceable compliance verification reports, so that the record compliance risk itself, as a heterogeneous input of comprehensive risk, also has interpretability. Attached Figure Description

[0008] Figure 1 This is a schematic diagram of the architecture of the system of the present invention in an offline intranet environment; Figure 2 A data flow diagram is created by combining basic case data, time-limit warning risks, and record compliance risks into a comprehensive case risk analysis. Figure 3 A flowchart for summarizing comprehensive case risks and updating event triggers; Figure 4 A flowchart for calculating the certainty of the case handling period; Figure 5 A flowchart for the sub-process of verifying the certainty and compliance of written records; Figure 6 This is a schematic diagram illustrating the interconnected display of comprehensive risks across multiple views, including case ledgers, case details, and report ledgers. Detailed Implementation

[0009] The present invention will be further described below with reference to the accompanying drawings and embodiments. The module described in the present invention is a logical module divided according to function, which can be carried by the same application deployed on the local machine of the public security intranet and does not depend on external network services.

[0010] Example 1 (Main Method Flow) A certain law enforcement case handling quality support system uses a unified case identifier to associate cases, delay records, time limits not included in the calculation, transcripts and their compliance verification reports and operation logs, and stores all data in an intranet local database that is physically isolated from the Internet.

[0011] Deterministic Calculation of Time Limit (Category 1 Risk): The time limit calculation module calculates according to configurable rules (e.g., basic processing time limit of 30 days, maximum of 2 extensions, 30 days for each extension, critical expiration threshold of 3 days, and warning expiration threshold of 10 days): The total time limit equals the basic processing time limit plus the product of the number of extensions and the number of days for each extension; the expiration date equals the case filing date plus the total time limit plus the number of days not included in the time limit; the remaining days equals the expiration date minus the current date; and the time limit warning risk is determined deterministically based on the processing status and the remaining days—if the processing status is "completed," it is not included in the near-expiration warning statistics; if the remaining days are less than zero, it is overdue; if the remaining days are not greater than the critical threshold, it is critically expired; if they are not greater than the warning threshold, it is warning expiration; otherwise, it is in normal processing; thus, overdue or critically expired is mapped to high risk, warning expiration is mapped to medium risk, and normal processing or completed is mapped to low risk. The number of extensions is limited to no more than the maximum number of extensions.

[0012] Record Certainty Verification (Category 2 Risk): The record compliance verification module performs certainty verification on the record texts associated with the case. First, it limits the length of the input text and blocks sensitive markers. Then, it identifies the record type (administrative inquiry, criminal interrogation, witness inquiry, victim inquiry, unknown type). Next, it performs line-by-line structured parsing of the title, start and end times, location, case handlers, recorder, person being questioned or interrogated, identity information, notification of rights and obligations, question and answer content, signature and fingerprint confirmation, template placeholders, and special procedure fields such as summons, minors, translation, and simultaneous audio and video recording. It also identifies the sequential relationship between elements (e.g., whether the notification of rights and obligations appears after the factual questions and answers, the order of signature dates, and the order of start and end times). During the parsing and matching process, it identifies negative contexts, such as using negative word patterns to exclude statements in negative contexts like "no refusal to sign" or "no further refusal to sign" from the "refusal to sign" related question items, avoiding false alarms. The system matches applicable rule subsets based on record type, and a centrally maintained deterministic compliance rule library identifies problem items in the parsing results. Rule matching does not rely on machine learning models or large language models; the matching chain is traceable. Each matched problem item is associated with its triggering rule's rule identifier, problem name, risk level, legal basis, and handling recommendations. Each rule in the deterministic compliance rule library is pre-configured with rule category, checkpoints, legal basis, and handling recommendations. Rule categories and checkpoints are used for rule organization and maintenance and can be traced back via the rule identifier. Matched problem items are sorted by risk level (high, medium, low), generating a priority handling list, a concise report, and a full report. The user-facing report hides internal rule codes and only displays understandable problem names and handling recommendations. The record compliance risk level is the highest risk level among its problem items. A case can be associated with multiple records; the highest risk level from the latest compliance verification reports of each record is taken as the record compliance risk level for that case.

[0013] Comprehensive Risk Synthesis: The risk synthesis module performs comprehensive risk synthesis according to preset synthesis rules. Since the deadline warning risk originates from the deterministic date calculations of structured case deadline data, and the transcript compliance risk originates from the deterministic rule matching of unstructured transcript texts, the data sources, data types, and calculation methods of these two types are heterogeneous and lack comparable unified dimensions, making direct numerical weighted summation impossible. Therefore, the two types of risks are first mapped to the same three-level risk scale consisting of high, medium, and low risk to achieve heterogeneous normalization. Then, the higher of the two is taken as the comprehensive risk level of the case, i.e., deterministic shortcoming synthesis. The technical rationale for using the higher value instead of weighted summation is that the dimensions of the two types of risks are incomparable. Weighted summation requires the introduction of weight coefficients that lack objective basis, require training, or are manually calibrated, and may dilute and mask the high risk in one dimension by the low risk in another. Taking the higher value ensures that when any dimension reaches high risk, the comprehensive risk of the case is high risk. The calculation is deterministic, can be reviewed item by item, and requires no training or calibration, meeting the practical requirement of "shortcomings are risks" in case handling. While synthesizing, a structured comprehensive risk text is generated, such as "Comprehensive high risk; 2 days remaining in the deadline, high risk; medium risk in the record; 1 record pending verification"; when there are records in a case that have not yet obtained a compliance verification report, the case is marked as having records pending verification.

[0014] Linked Display and Event Refresh: The comprehensive risk and traceable entry points pointing to the deadline calculation results and compliance verification reports are displayed in multiple views of the case ledger, case details, and report ledger. When a case is added, edited, closed, resumed, softly deleted, imported in batches, or a new record is added or a compliance verification report is generated, the system automatically recalculates and writes or updates the comprehensive risk of the case according to the logic bound to the business semantics. When the case is closed, it is removed from the near-expiration warning statistics but its historical compliance verification report is retained. When the case is resumed, it is re-included in the near-expiration warning statistics and the comprehensive risk is refreshed. When the case is softly deleted, the comprehensive risk summary record of the case is cleared and recalculated after resumption. When a new record is added or a compliance verification report is generated, the risk synthesis module pulls the latest compliance verification report risk of the record and resynthesizes it. When a batch is imported, the deadline for each imported case is automatically calculated and written to the import operation log. The refresh criteria for each of the aforementioned events are different and each is semantically linked to the case handling process. This differs from the common practice in general information systems where all indicators are recalculated indiscriminately once the status changes. Upon completion, only the inclusion of the case in the near-term warning statistics is changed, and its historical compliance verification report is retained without deletion. Soft deletion clears the comprehensive risk summary record but retains recoverability, allowing for recalculation after recovery. New records or generated reports only retrieve the risk level from the latest compliance verification report for recombining, thus maintaining consistency in the linked summary with minimal and semantically correct updates. The entire process is completed within the local database on the internal network; case data, record texts, and reports are not sent to external networks.

[0015] Example 2 (Data Example) Input Case: Case number is a demonstration case number, the handling unit is a demonstration police station, the case filing date is 2026-06-01, the number of extensions is 0, the number of days not included in the time limit is 0, and the case is in progress. Time Limit Calculation Output: Total time limit 30 days, expiry date 2026-07-01, remaining days calculated based on the current date, and time limit warning risk changes with the remaining days. This case is associated with an administrative inquiry record, and the certainty check hits the medium-risk issue item "the location of the fact is too general," indicating that the compliance risk of this record is medium risk. If the time limit approaches its critical expiry date (high risk), the risk synthesis output is a comprehensive high risk for the case, which is displayed and traceable in the case ledger, case details, and report ledger. When the case is closed, it will no longer be included in the near-expiration warning statistics, but the historical compliance check report will be retained; after the case is resumed, it will be re-included and the comprehensive risk will be refreshed; after being soft-deleted, its comprehensive risk summary record will be cleared, and it will be recalculated after restoration.

[0016] Example 3 (Negative Context and Abnormal Isolation) When the transcript states "the person being questioned did not refuse to sign," the signature refusal rules are not triggered due to the negative context recognition, thus avoiding false alarms. Situations such as unreplaced template placeholders, lack of reasoned documentation during administrative inquiries, and lack of notification of defense or legal assistance during first criminal interrogations are addressed with corresponding issues and checkpoints and legal basis output. When an anomaly occurs in the local database during comprehensive risk recalculation, the system isolates and records the failed refresh of a single case, without affecting the refresh of other cases; batch refreshes return the total number of cases and the number of successful refreshes, ensuring the robustness of the linked aggregation.

[0017] Example 4 (System, Storage Media, and Computer Program Products) The above method can be implemented by a system deployed on the public security intranet. This system includes an association module, a time limit calculation module, a record compliance verification module, a risk synthesis module, and an event refresh module. It may also include a memory and one or more processors. When the computer instructions in the memory are executed by the processor, the system performs the above method. Alternatively, the above method can be implemented by a computer-readable storage medium storing a computer program, or by a computer program product containing a computer program or instructions. When run on an electronic device, it causes the electronic device to perform the steps of the above method. The system is deployed as a self-contained offline runtime package. The offline runtime package includes application artifacts, a portable runtime, local database files, and startup, backup, and data import / export scripts. Ordinary case-handling computers can access the system only through a browser without needing to install a server-side runtime environment.

Claims

1. A method for linking and summarizing the risks of case handling time limits and record compliance in an intranet environment, applied to a law enforcement case handling quality support system deployed on a public security intranet physically isolated from the Internet, characterized in that... include: Establish a unified case identifier to link basic case data, case handling period calculation results, transcripts and their compliance verification reports, and operation logs; The time limit calculation module determines the total time limit, due date, and remaining days of the case based on the case filing date, basic processing time limit, number of extensions and the number of days per extension, days not included in the time limit, and processing status. The processing status and remaining days then determine the first type of risk, which is a time limit warning risk. The record compliance verification module performs rule matching on the record text associated with the case using a deterministic compliance rule base, determining the second type of risk, which is also a record compliance risk. Each matched issue carries the rule identifier of the triggered rule, and the rule identifier is used to trace back to the rule category, checkpoint, legal basis, and processing suggestions in the deterministic compliance rule base, thus forming a rule matching chain that can be reviewed item by item. Furthermore, the internal rule codes, such as the rule identifier, are hidden in the user-facing output. The compliance risk of the transcript is not determined based on the uncertain output of machine learning models or large language models; the risk synthesis module combines the first type of risk and the second type of risk into a comprehensive risk level for the case according to a preset synthesis rule. The first type of risk comes from deterministic date calculations on structured case timeline data, and the second type of risk comes from deterministic rule matching on unstructured transcript natural language text. The data sources, data types, and calculation methods of the first type of risk and the second type of risk are heterogeneous and their dimensions are incomparable. The preset synthesis rule is as follows: the first type of risk and the second type of risk are each mapped to a unified three-level risk level scale including high risk, medium risk, and low risk, and the higher risk level is taken as the comprehensive risk level for the case. In response to events such as case closure, resumption of processing, soft deletion, addition of records, or generation of compliance verification reports, the comprehensive risk level of the case is automatically recalculated and refreshed; wherein, the basic case data, record texts, and compliance verification reports are all stored and processed locally on the intranet and are not sent to external networks.

2. The method according to claim 1, characterized in that, The deadline warning risk is determined based on the processing status and the remaining days as one of the following: when the processing status is completed, it is considered completed and not included in the near-expiration warning statistics; when the remaining days are less than zero, it is considered overdue; when the remaining days are greater than or equal to zero and not greater than the first threshold, it is considered critically due; when the remaining days are greater than the first threshold and not greater than the second threshold, it is considered a warning due date. Otherwise, it is being processed normally; and the time limit warning risk is mapped as high risk when it has exceeded or is about to expire, medium risk when it is warning expired, and low risk when it is being processed normally or has been completed; the total time limit is equal to the sum of the product of the basic processing time limit, the number of extensions, and the number of days for each extension, the expiration date is equal to the case filing date plus the sum of the total time limit and the number of days not included in the time limit, the remaining days are equal to the expiration date minus the current date, and the number of extensions is limited to no more than the preset maximum number of extensions.

3. The method according to claim 1, characterized in that, The second type of risk, namely the record compliance risk, is represented by three levels: high risk, medium risk, and low risk, and is on the same three-level risk scale as the risk level mapped by the first type of risk. When the case involves multiple records, the record compliance risk level is the highest among the risk levels in the latest compliance verification reports of each record in the case. The comprehensive risk level is the higher of the risk level mapped by the first type of risk and the record compliance risk level, rather than a weighted sum of the two types of risks.

4. The method according to claim 1, characterized in that, When the comprehensive risk level of the case is refreshed, a structured comprehensive risk text is generated. The comprehensive risk text includes the comprehensive risk level, the remaining days when the remaining days are non-negative or the overdue days when they are negative and their corresponding risk levels, the compliance risk level of the records, and the number of records to be verified. When there are records in the case that have not yet obtained a compliance verification report risk level, the case is marked as having records to be verified. The comprehensive risk level and the traceable entry points pointing to the case handling period calculation results and the compliance verification report are displayed together in the case ledger view, case details view and report ledger view.

5. The method according to claim 1, characterized in that, The system automatically recalculates and refreshes the comprehensive risk level of a case in response to events such as case completion, resumption of handling, soft deletion, addition of records, or generation of compliance verification reports. This includes: case completion events removing the case from the near-term warning statistics while retaining its historical compliance verification reports; resumption of handling events re-entering the case into the near-term warning statistics and refreshing the comprehensive risk level; soft deletion events clearing the case's comprehensive risk summary record and recalculating it after the case is resumed; and adding records or generating compliance verification reports causing the risk synthesis module to retrieve the risk level from the latest compliance verification report of the record and resynthesize the case's comprehensive risk level. Furthermore, in response to batch import events, the system automatically calculates the case handling period for each imported case, refreshes its comprehensive risk level, and writes it to the import operation log.

6. The method according to claim 1, characterized in that, The record compliance verification module determines the second type of risk by performing rule matching on the record text associated with the case using a deterministic compliance rule base. This includes: identifying the record type of the record text, which includes at least administrative inquiry, criminal interrogation, witness inquiry, victim inquiry, and unknown types; matching the corresponding rule subset according to the record type; performing line-by-line structured parsing of the record text before rule matching, extracting record elements including title, start and end time, location, case handler, recorder, person being questioned or interrogated, identity information, notification of rights and obligations, question and answer content, signature and fingerprint confirmation, as well as template placeholders and special procedure fields; and identifying the sequential relationship between the record elements, so that rule matching includes a deterministic determination of the sequential relationship between elements.

7. The method according to claim 6, characterized in that, Negative context recognition is performed in the trigger determination of the rule hit: the transcript text contains statements that refuse to sign, refuse to fingerprint, or do not occur, and statements in negative context are excluded from the hit of the corresponding question item, thereby distinguishing between positive and negative statements to determine whether the corresponding rule is hit.

8. The method according to claim 6, characterized in that, After the rule hits the identified problem items, it associates each identified problem item with a problem name, risk level, legal basis, and handling suggestions, and sorts the problem items according to risk level to generate a priority handling list and a graded compliance verification report. The compliance risk level of the record is determined by the highest risk level among the issue items; wherein, each rule in the deterministic compliance rule base is pre-configured with its rule category, checkpoint, legal basis, and handling suggestions; the rule hit chain traceability includes: recording the rule identifier, issue name, risk level, legal basis, and handling suggestions of the triggering rule for each hit issue item in the data layer; the rule category and checkpoint are obtained by tracing back to the rules in the deterministic compliance rule base through the rule identifier; in the user-facing hierarchical compliance verification report layer, the issue name, risk level, and handling suggestions are displayed according to the risk level, and the internal rule code such as the rule identifier is hidden.

9. A system for linking and summarizing case handling time limit risks and record compliance risks in an intranet environment, deployed on a public security intranet physically isolated from the Internet, characterized in that... The system includes a local database, a storage device, one or more processors, and the following modules executed by the processors: an association module for establishing a unified case identifier and associating basic case data, case handling period calculation results, transcript documents and their compliance verification reports, and operation logs; a period calculation module for deterministically calculating the total period, due date, and remaining days based on the case's filing date, basic handling period, number of extensions and the number of days for each extension, days not included in the period, and handling status, and determining the period warning risk as the first type of risk; and a transcript compliance verification module for deterministically matching the transcript text associated with the case using a deterministic compliance rule base to determine the transcript compliance risk as the second type of risk. The risk synthesis module is used to synthesize the first type of risk and the second type of risk into a comprehensive risk level for the case according to a preset synthesis rule. The event refresh module is used to automatically recalculate and refresh the comprehensive risk level of a case in response to events such as case closure, resumption of processing, soft deletion, addition of records, or generation of compliance verification reports. The memory stores computer instructions, which, when executed by one or more processors, cause the system to perform the method as described in any one of claims 1 to 8. The system is deployed on the intranet as a self-contained offline runtime package, which includes application artifacts, a portable runtime, local database files, and startup, backup, and data import / export scripts. Ordinary case-handling computers can access the system only through a browser without needing to install a server-side runtime environment.

10. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instructions are run on an electronic device, the electronic device causes the electronic device to perform the steps of the method as described in any one of claims 1 to 8.

Images