Intelligent legal affairs collaborative processing system based on legal knowledge graph

Abstract

The application is specifically a wisdom legal affairs collaborative processing system based on a legal knowledge graph, relates to the technical field of legal affairs informatization, and comprises a legal information ontology and space-time label module, a three-dimensional dynamic strategy matrix management module, a real-time view synthesis module, a process state synchronization and triggering module, and an audit and strategy tracing module.In the application, the legal information ontology and space-time label module defines standardized types and assigns situational labels to collaborative information, clarifies information association, and lays a standard data foundation; the three-dimensional dynamic strategy matrix management module constructs a dynamic rule engine associated with roles, stages and information attributes, supports flexible definition and adjustment of permissions, can avoid leakage of sensitive information or insufficient information access, a visual configuration device can reduce the threshold of rule control, significantly improves the accuracy and security of legal affairs permission control, and guarantees compliance of information use.

Description

Technical Field

[0001] This invention relates to the field of legal information technology, and in particular to a smart legal collaborative processing system based on legal knowledge graphs. Background Technology

[0002] With the increasing complexity and diversification of legal affairs, traditional legal collaboration models are no longer sufficient to meet the demands for efficient, accurate, and compliant processing.

[0003] Current legal collaboration systems generally suffer from the following problems: Information management is chaotic, lacking a unified standardized definition of information, and different types of legal documents, evidence materials and other information are stored in a mess, making it difficult to achieve efficient retrieval and association; Rigid access control, often using a fixed role-based access control model, cannot dynamically adjust permissions according to changes in the stage of a case, which can easily lead to the leakage of sensitive information or insufficient access to information. The process synchronization is lagging, and the case stages in the system are not synchronized with the actual legal process, making it impossible to drive the dynamic adjustment of permissions and views in a timely manner.

[0004] Legal knowledge graphs, as an effective tool for knowledge organization and management, can clearly display the semantic relationships between legal information. Applying legal knowledge graphs to legal collaborative processing systems is expected to address the shortcomings of the aforementioned traditional systems.

[0005] Therefore, a smart legal collaborative processing system based on legal knowledge graph is proposed to realize standardized management of legal information, dynamic adaptation of permissions, personalized generation of views, real-time synchronization of processes, and full-process audit traceability, thereby improving the intelligence level and compliance of legal collaborative processing. Summary of the Invention

[0006] The purpose of this invention is to propose an intelligent legal collaborative processing system based on legal knowledge graphs in order to solve the above-mentioned problems.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: A smart legal collaborative processing system based on legal knowledge graphs includes: The legal information ontology and spatiotemporal label module is configured to standardize the type definition and assign contextual spatiotemporal labels to legal collaborative information, laying the data foundation for the generation of dynamic views; The three-dimensional dynamic strategy matrix management module is configured as a dynamic rule engine to build relationships between roles, stages, and information attributes, enabling the definition, reasoning, and stage-based adaptation of permissions. The real-time view composition module is configured to capture a three-dimensional context of user role, case stage, and current intent, and assemble a personalized interface that meets permissions and requirements in real time. The process status synchronization and triggering module is configured to synchronize the actual process stages of the case and publish global events, triggering batch re-evaluation of policy rules and driving the view to change dynamically with the process. The audit and strategy traceability module is configured to record full-dimensional, tamper-proof operation logs, providing simulation analysis of the impact before strategy modifications, ensuring the compliance of system operations and the rationality of strategies.

[0008] Preferably, the legal information ontology and spatiotemporal tagging module specifically includes: Define the types and attribute relationships of all information objects in legal collaboration to form a dynamically extensible lightweight ontology; The system adopts a hierarchical classification structure, with the top level divided into four major categories: litigation documents, evidence materials, internal collaborative information, and notices from judicial organs. Each major category is further subdivided into specific types. The semantic relationships and constraints between information objects are defined, and are divided into three core relationships: support relationships, adversarial relationships, and evolutionary relationships. Standardized type labels are assigned to every document and every piece of data within the system, and these labels contain hierarchical type paths. Attach dynamically adjustable contextual tags to each piece of information; Four-dimensional structured tags are used: Generation phase: Specific litigation process nodes where marker information is generated; Target Phase: The specific phase in which information should be disclosed or used according to law or policy, and the associated triggering conditions; Strategic security classification: Based on the sensitivity of the information and the scope of use, it is divided into four levels, including public, public to the court, confidential to our party, and absolutely confidential; Information granularity: The scope of application of tagging, divided into document level, paragraph level, and field level; Tag assignment supports batch templated and personalized adjustments; Each output message carries a structured label: [type, generation stage, target stage, policy security level, information granularity].

[0009] Preferably, the three-dimensional dynamic strategy matrix management module specifically includes: Define the dynamic permission mapping relationship between Role, Stage, and Info attribute; The core of the model is the rule function: Permission=F(Role,Stage,Info_Type,Info_Generate_Stage,Info_Target_Stage,Info_Security_Level,Info_Granularity); In this context, Role corresponds to a character role; Stage corresponds to a phase. The Info attribute contains: Information type (Info_Type), information generation stage (Info_Generate_Stage), information target stage (Info_Target_Stage), information security level (Info_Security_Level), and information granularity (Info_Granularity); The function outputs a set of operation permissions; Role: A hierarchical role system is adopted, divided into core roles and sub-roles: Stage: Consistent with the generation stage and target stage of the legal information ontology and spatiotemporal tagging module; Information Attribute Info: The information ontology tags and spatiotemporal tags that integrate the legal information ontology and spatiotemporal tag module include information type, generation stage, target stage, policy security level, and information granularity; Define rule priorities; when multiple rules conflict, the highest-priority rule will be executed. The final output is a policy rule base containing all the rules.

[0010] Preferably, the system further includes a visual policy configurator submodule: Provides preset rule templates that cover the core rules for common case types; When configuring rules, rule conflicts are detected in real time, and the currently configured rules are compared with the existing rules in the rule base. The addition, modification, and deletion of rules shall take effect only after approval. Approved and effective policy rules are automatically synchronized to the policy rule library and become the rule basis for the view composition engine.

[0011] Preferably, the real-time view synthesis module specifically includes: Input information collection: User Identity Role: Obtained through the system's identity authentication module, it includes the user's core role and sub-roles, and supports multiple roles overlapping; The current official stage of the case: obtained from the process status synchronization and triggering module, is the latest official stage of the case. Only the current stage is used as the core parameter for view composition. The current user intent is inferred from the user's actions, or the user can manually select the current task intent in the system interface. Contextual validation: Verify whether the user's role is a participant in the case; Verify whether the current stage of the case is a valid stage; Verify whether the user's intent matches the current stage; If the verification fails, prompt the user and adjust the context parameters; The final output is a triple (R, S, Intent), where R is the user's set of roles, S is the current case stage, and Intent is the user's current intent.

[0012] Preferably, the method further includes: Using a multi-condition query, the logical expression is to find all information objects that meet the following conditions: The current user role, within the current stage, has viewing or higher permissions for this information object; The target stage of this information object is less than or equal to the current stage, or the generation stage is highly relevant to the current user intent. The information granularity of this information object conforms to the current user's permission scope; The information object was not marked as invalid; After the query results are returned, the system assembles the view from three dimensions based on the permission rules and information attributes, including display field control, relationship display, and operation button control; Output a dynamic view that includes a list of information that the user has the right to view, the fields displayed for the information, the relationships between the information, and the action buttons that the user can perform.

[0013] Preferably, the process status synchronization and triggering module specifically includes: It interfaces with external judicial systems and synchronizes in real time. For cases that cannot be integrated with external judicial systems, a manual confirmation process is provided: After a case is updated, a notification of the phase change will be sent to all parties involved in the case. After the phase update operation is completed, the operation is published as a global event for re-evaluation. The latest official stage of the case has been finalized. Real-time monitoring of process status synchronization and global events for stage changes published by the triggering module; when an event is captured, automatic permission re-evaluation of the process is initiated. Based on the new case stage, the policy rule base is invoked to recalculate the permission set for each role for the selected information objects. The re-evaluated permission results are updated in the information permission database, and the updated information permission set is output and stored in the information permission database.

[0014] Preferably, the audit and strategy tracing module specifically includes: The log uses a structured format, and each log entry contains the following core fields: [Log ID, timestamp, user ID, user role, case number, case stage, operation type, information object ID, information object type, triggered policy rule ID, operation result, client IP, operation terminal]; The triggered policy rule ID field is used to associate the permission rule on which the operation is based; The log records cover all core system operations, including information operations, rule operations, stage operations, and view operations.

[0015] Preferably, the method further includes: Simulation scenario selection: Supports historical stage simulation and future stage prediction; Display a comparison of the logical expressions of the original and new rules, and present the changes to the rules; Simulate the changes in information object permissions after the new rules take effect, and statistically analyze the number and types of information objects with increased permissions, the number and types of information objects with decreased permissions, and the changes in the scope of information exposure.

[0016] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: 1. This invention standardizes the type definition and contextualized label assignment of collaborative information through a legal information ontology and spatiotemporal tag module, clarifies information relationships, and lays the foundation for standardized data. Combined with a three-dimensional dynamic strategy matrix management module, it constructs a dynamic rule engine that associates roles, stages, and information attributes, supports flexible definition and adjustment of permissions, avoids leakage of sensitive information or insufficient information access, and lowers the threshold for rule control through a visual configurator, significantly improving the accuracy and security of legal authority control and ensuring compliance in information use.

[0017] 2. This invention captures the three-in-one context of user role, case stage, and current intent through a real-time view synthesis module, assembling a personalized interface in real time, reducing information overload, and helping users quickly obtain key information; through a process status synchronization and triggering module, it realizes real-time synchronization between system stages and real legal processes, publishes global event trigger permissions for re-evaluation, drives the view to update dynamically with the process, solves the problems of fixed views and lagging process synchronization in traditional systems, ensures accurate and timely information display, and greatly improves the efficiency of legal collaborative processing. Attached Figure Description

[0018] Further details, features, and advantages of this application are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which: Figure 1 This is a system structure diagram of the present invention. Detailed Implementation

[0019] Several embodiments of this application will now be described in more detail with reference to the accompanying drawings to enable those skilled in the art to implement this application. This application may be embodied in many different forms and for various purposes and should not be limited to the embodiments set forth herein. These embodiments are provided to make this application thorough and complete, and to fully convey the scope of this application to those skilled in the art. The embodiments described do not limit this application.

[0020] Unless otherwise defined, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. It will be further understood that terms such as those defined in commonly used dictionaries shall be interpreted as having a meaning consistent with their meaning in the relevant field and / or the context of this specification, and shall not be interpreted in an idealized or overly formal sense unless expressly defined herein.

[0021] Example 1: Its specific implementation method is combined with the appendix Figure 1 Please provide a detailed explanation.

[0022] Appendix Figure 1 The block diagram of the intelligent legal collaborative processing system based on legal knowledge graph provided in the embodiments of the present invention shows the connection relationship between the legal information ontology and spatiotemporal tag module and the audit and strategy tracing module, and marks the main functional interaction flow of each module.

[0023] In this embodiment, it includes: The legal information ontology and spatiotemporal label module is configured to standardize the type definition and assign contextual spatiotemporal labels to legal collaborative information, laying the data foundation for the generation of dynamic views; Specifically, it includes: Legal Information Ontology Submodule: Define the types and attribute relationships of all information objects in legal collaboration to form a dynamically extensible lightweight ontology. This ontology does not rely on a complex semantic web framework, but is customized based on legal practice scenarios to ensure that it meets standardization requirements while having efficient engineering feasibility.

[0024] Information types: A hierarchical classification structure is adopted, with the top level divided into four major categories: litigation documents, evidence materials, internal collaborative information, and notices from judicial organs. Each major category is further subdivided into specific types. For example: Litigation documents: complaint, answer, appeal, attorney's statement, cross-examination opinions, draft judgment, etc.; Evidence materials: documentary evidence (contracts, invoices, official letters), physical evidence (samples of infringing products and equipment), audio-visual materials (audio recordings, video recordings, electronic data), witness testimonies, expert opinions, etc.; Internal collaborative information: internal discussion minutes, case strategy analysis, client communication records, pre-trial preparation outlines, etc.; Notifications from judicial authorities: court summons, notices of evidence submission, orders for exchange of evidence, court hearing announcements, rulings, etc.

[0025] Attribute relationships: Define the semantic associations and constraints between information objects, divided into three core categories, including: Supporting relationships: such as evidence-claim support relationships (one piece of documentary evidence can support multiple claims in a case, one claim can be associated with multiple pieces of evidence, and the supporting relationships can be weighted for priority ranking). Adversarial relationship: such as the document-rebuttal relationship (the answer is against the complaint, the cross-examination opinions are against the opposing party's evidence, the adversarial relationship has a two-way connection, and the specific content of the rebuttal can be traced back to the target). Historical relationships: such as document-version historical relationships (there is a succession relationship between different versions of the same document; new versions can be marked with modified content, while old versions cannot be deleted, but can only be marked as invalid).

[0026] Implementation mechanism: The ontology library adopts configuration-based management, which allows system administrators to add exclusive information types based on new case types (such as intellectual property litigation and labor dispute litigation), or add attribute relationships (such as evidence-legal citation relationship) according to practical needs.

[0027] Output: Label each document and each piece of data in the system with a standardized type label. The label contains a hierarchical type path (such as evidence materials - documentary evidence - contract documents) to ensure the uniqueness and identifiability of information types.

[0028] Information Spatiotemporal Tag Atomization Submodule: Function: Attach dynamic and adjustable contextual tags to each piece of information (or information block, such as a page in a piece of evidence or a paragraph in a proxy statement), so that the information has contextual intelligence regarding when it was generated, when it was used, and who can access it; Tagging System: Employs a four-dimensional structured tag system, where each tag supports further subdivision and expansion. Generation stage: This stage marks specific litigation process nodes generated by the information, supporting multi-level subdivisions. For example: Pre-litigation preparation stage: client contact, case analysis, evidence collection, and strategy development; First instance stage: case filing, evidence exchange, pre-trial conference, trial (court investigation, court debate), and pronouncement of judgment; Execution phase: application for execution, property investigation and control, execution settlement, and compulsory execution.

[0029] The generation stage assignment method is automatic extraction and manual confirmation: the system can automatically match the stage based on the document creation time and associated operation records, and also supports manual adjustment by the user to ensure the accuracy of the stage marking.

[0030] Target Phase: Marks the specific phase at which information should be disclosed or used according to law or policy, and associates it with triggering conditions. For example: Statutory trigger: The target stage of evidence is the exchange of evidence, and the triggering condition is the issuance of a notice of evidence submission by the court. Strategy Trigger: The target stage for pre-trial evidence gathering is court investigation, and the trigger condition is that the lead attorney manually confirms the submission. Internal use: The target stage of the case strategy analysis is for internal use throughout the entire process, with no external disclosure triggering conditions.

[0031] Security Policy Classification: Based on information sensitivity and scope of use, it is divided into four levels. Each level defines clear access subjects and operation permission boundaries, including: Public: Applicable to effective judgments, court announcements, etc. Access is available to all individuals, with permissions including viewing, downloading, and sharing. Open to the court: Applicable to evidence, statements of defense, etc. that are required to be presented to the court and the opposing party in accordance with the law. Access is available to judges, personnel of one's own side, and personnel of the opposing party. Operation permissions are to view and download, but not to view internal annotations. This is confidential for internal use only: it is suitable for internal discussion transcripts, case strategy analyses, etc. Access is limited to our lawyers and clients, and access permissions include viewing, editing, and linking. Absolute confidentiality: Applicable to clients' core business secrets, core attack and defense strategies for cases, etc. Access is limited to specific key personnel (such as the lead lawyer or the client's legal representative). Operation permissions are limited to viewing only; editing, downloading, and sharing are not allowed.

[0032] Information granularity: The scope of application of the tag is divided into document level, paragraph level, and field level. For example, the customer privacy information field in the customer communication record can be marked as absolutely confidential, while the entire document can be marked as confidential by this party.

[0033] The assignment of tags supports batch templates and personalized adjustments. For similar information (such as documents collected in the same batch), the same tags can be applied in batches, and for special information, the tag content can be adjusted individually. Meanwhile, the tags support dynamic updates. When the case progresses or the strategy is adjusted, the target stage or strategy confidentiality level of the information can be modified. The update operation must be logged and go through an approval process. Each output message (or message block) carries a structured tag: [type, generation stage, target stage, policy security level, message granularity]. This tag, together with the message ontology tag, constitutes a complete structured description of the message.

[0034] The three-dimensional dynamic strategy matrix management module is configured as a dynamic rule engine to build relationships between roles, stages, and information attributes, enabling the definition, reasoning, and stage-based adaptation of permissions. It is the core of the system's rules. Specifically, it includes: 3D Strategy Model Submodule: Function: Defines the dynamic permission mapping relationship between Role, Stage, and Info, and builds a policy rule system that can cover the entire process, all roles, and all information types; The core of the model is a scalable rule function: Permission=F(Role,Stage,Info_Type,Info_Generate_Stage,Info_Target_Stage,Info_Security_Level,Info_Granularity); In this context, Role corresponds to a character role; Stage corresponds to a phase. The Info attribute contains: Information type (Info_Type), information generation stage (Info_Generate_Stage), information target stage (Info_Target_Stage), information security level (Info_Security_Level), and information granularity (Info_Granularity); F is an abstract identifier for the rule function, which is an encapsulation of the permission calculation logic and is used for the core processing logic of the rule engine to calculate permissions. Specifically, F is essentially a logical processing set that integrates matching conditions, priority rules, and permission set mapping relationships for all roles, stages, and information attributes in the policy rule base. It receives parameters from seven dimensions: Role, Stage, and detailed information attribute items, and outputs permission results through the following core logic: Match the corresponding rule conditions in the matching strategy rule base; Valid rules are selected based on rule priority; Based on the definition of valid rules, generate the corresponding set of operation permissions (i.e., Permissions).

[0035] The function takes seven parameters as input and outputs a specific set of operation permissions, supporting multiple condition combinations and priority sorting. Dimensional breakdown: Role: A hierarchical role system is adopted, divided into core roles and sub-roles: Key roles: Chinese lawyer, opposing lawyer, judge, client, court clerk, expert witness, etc. Roles are further subdivided: Under our lawyers, there are lead lawyers, co-lead lawyers, and trainee lawyers; under the client, there are legal representatives, authorized agents, and general contacts, etc.

[0036] Roles can be added dynamically, and exclusive roles can be added based on the type of parties involved in the case (such as arbitrators in arbitration cases or agents of administrative agencies in administrative litigation).

[0037] Stage: Consistent with the generation stage and target stage of the legal information ontology and spatiotemporal tag module, it supports custom expansion and can add exclusive stages according to special case processes (such as bankruptcy liquidation and public interest litigation); Information Attribute Info: The information ontology tags and spatiotemporal tags of the legal information ontology and spatiotemporal tag module are integrated, including five core attributes: information type, generation stage, target stage, strategy security level, and information granularity. Define four levels of rule priority to ensure the uniqueness and consistency of rule execution, including: Legal rules: based on legally prescribed authority constraints (such as a judge's right to view all submitted evidence at any stage); Core strategy rules: Access restrictions based on the core attack and defense strategies of the case (e.g., absolutely confidential information can only be viewed by core personnel); General policy rules: Standard access restrictions applicable to all cases (e.g., opposing counsel cannot view the opposing counsel's internal policies); Custom rules: Personalized permission constraints for preset cases and preset information; When multiple rules conflict, the system automatically executes the highest-level rule according to its priority. The final output is a policy rule library containing all rules. The rules are stored as standardized logical expressions and can be called by the visual configurator and queried by the view composition engine.

[0038] Example of a rule (using evidence as an example): Example 1 (Statutory Rule): When Stage = any stage, Role = judge, Info_Type = evidence and Info_Target_Stage <= current Stage, Permission = {viewable, downloadable, viewable all annotations}, with a priority of level 1.

[0039] Example 2 (Core Strategy Rule): When Stage = Evidence Exchange, Role = Opposing Counsel, Info_Type = Evidence, and Info_Target_Stage = Evidence Exchange and Info_Security_Level = Open to Court, Permission = {Viewable, Downloadable, No Access to Internal Annotations}, with a priority level of 2.

[0040] Example 3 (General Strategy Rule): When Stage = Pre-trial Conference, Role = Our Trainee Attorney, Info_Type = Internal Strategy Analysis, Info_Security_Level = Our Confidentiality, Permission = {Viewable, No Editing, Can Be Linked to Case Claims}, the priority level is 3.

[0041] Example 4 (Stage Triggering Rule): When the Stage progresses from evidence exchange to court hearing, the system automatically triggers the rule to update the Permission of the judge's role in the Info_Type= evidence exchanged previously from pending to viewable. At the same time, the system adds the ability to submit cross-examination opinions to the lawyer's own side, with a priority level of 2.

[0042] It also includes a visual policy configurator submodule: Function: Allows system administrators (or case-leading lawyers) to configure, modify, and delete policy rules in a graphical and visual way using if-then statements, without writing code, thus lowering the technical threshold for rule configuration.

[0043] Core UI Functionality: Rule Template Library: Provides preset rule templates covering the core rules of common case types (such as trade secret infringement cases and contract dispute cases). Users can directly call the templates and fine-tune the parameters to improve configuration efficiency. Multi-condition combination configuration: Select role (supports multiple selections), stage (supports multiple selections), and information attributes (information type, policy security level, etc., supporting multiple condition combinations) through the drop-down menu to construct the if condition part of the rule; Permission set configuration: By selecting operation permissions (view, edit, comment, download, associate, print, share, view comments), the result part of the rule is constructed, which supports configuring different permissions for different information granularities (such as document level can be viewed, field level cannot be viewed); Rule conflict detection: When configuring rules, the system automatically detects rule conflicts in real time and compares the currently configured rules with the existing rules in the rule library. If there are overlapping roles, stages, information attributes and permission conflicts, the system will prompt the ID and content of the conflicting rules and provide priority adjustment suggestions. Strategy Package Management: Supports packaging a set of rules into a dedicated strategy package, which can be bound to specific case types, such as a strategy package for trade secret infringement cases or a strategy package for labor disputes. Strategy packages support importing, exporting, and reuse, and can be shared between different cases and different systems; The addition, modification, and deletion of rules require an approval process. The person configuring the rules must submit an application, and the rules will only take effect after the lead lawyer (or system administrator) approves them, thus ensuring the security and rationality of the rule configuration. Output: Approved and effective policy rules are automatically synchronized to the policy rule base and become the rule basis for the view composition engine.

[0044] The real-time view composition module is configured to capture the three-dimensional context of user role, case stage, and current intent, and assemble a personalized interface that meets permissions and requirements in real time. It is the core of the system's rendering. Specifically, it includes: It is the core of the system's rendering, responsible for assembling a personalized interface in real time and dynamically, matching the user's current role, case stage, and intent, the moment the user accesses the system. Its core value lies in ensuring that each user can only see the information they are authorized to view and need to view, thus avoiding information overload and the leakage of sensitive information.

[0045] Context-aware submodule: Function: Captures and confirms the current user's triadic context in real time to ensure that the input parameters for view composition are accurate.

[0046] Input information collection: User Identity Role: Obtained through the system's identity authentication module, including the user's core role and sub-roles (such as the client's lawyer - lead lawyer), and supports multiple role stacking (such as the user simultaneously having the roles of client's lawyer and authorized agent). The current official stage of the case: obtained from the process status synchronization and triggering module, is the latest official stage of the case (such as the evidence exchange stage). It supports viewing historical stages, but only the current stage is used as the core parameter for view composition. User's current intent (Intent): Obtained through two methods: intelligent inference and manual selection. Intelligent inference: Based on the user's operation behavior (such as clicking the evidence preparation menu or searching for keywords for cross-examination opinions), the system intelligently infers the user's current intention (such as evidence preparation for pre-trial proceedings). Manual selection: Users can manually select the current work intent (such as customer communication strategy formulation) in the system interface. Manually selected intents have higher priority than intelligent inference. Contextual validation: The system performs cross-validation on the collected three-dimensional context. Verify whether the user's role is a participant in the case; Verify whether the current stage of the case is a valid stage; Verify whether the user's intent matches the current stage (e.g., the execution stage does not support the intent for pre-trial preparation). If the verification fails, the system will prompt the user and adjust the context parameters (such as the default intent being case viewing). The final output is a definite triple (R, S, Intent), where R is the user's set of roles, S is the current case stage, and Intent is the user's current intent. This triple serves as the core input parameter for view composition.

[0047] Dynamic query and assembly submodule: Function: Based on the triple (R,S,Intent), request permission rules from the three-dimensional dynamic strategy matrix management module, initiate a composite query from the legal information ontology and spatiotemporal label module and the database, and assemble a personalized view in real time based on the query results and permission rules.

[0048] Compound query logic: Employs precise queries with multiple overlay conditions. The logical expression is to find all information objects that satisfy the following conditions: The current user role, within the current stage, has viewing or higher permissions for this information object; The target stage of this information object is less than or equal to the current stage, or the generation stage is highly relevant to the current user intent (relevance ≥ 80%). The information granularity of this information object conforms to the current user's permission scope; The information object was not marked as invalid; To improve query efficiency, the system adopts index optimization and caching mechanisms: a dedicated index is created for the tag field of the information, and high-frequency query results (such as the view query results of the lawyers in the current stage) are cached. The cache validity period expires automatically as the stage changes or the rules are modified.

[0049] View assembly logic: After the query results are returned, the system assembles the view from three dimensions based on permission rules and information attributes, including: Display field control: Enables precise permission control at the field level, determining which fields of an information object can be displayed based on user roles and information policy security levels. For example: When the opposing lawyer reviews the evidence, only the evidence name, evidence type, and evidence summary fields are displayed; When our lead attorney reviews the same evidence, additional fields are displayed for the source of the evidence, the method of obtaining the evidence, the purpose of the evidence, internal annotations, and the supporting claims. When trainee lawyers review the same evidence, the internal annotation fields are not displayed.

[0050] The configuration of the displayed fields is bound to the policy rules and can be dynamically adjusted.

[0051] Relationship Display: Dynamically generates an information relationship view based on user role and current intent, helping users quickly understand the logical relationships between information. For example: Our lawyers intend to generate a relationship diagram of evidence, legal provisions, claims, and rebuttals when formulating strategies, demonstrating the degree to which the evidence supports the claims, the applicability of the legal provisions to the evidence, and the possible directions of the opposing party's rebuttals. The judge intends to generate a comparative view of the points of contention, evidence from both sides, and claims from both sides during the trial, presenting the evidence and claims of both sides on the same point of contention side by side to facilitate the trial. When clients want to understand the progress of a case, a flow view is generated showing the case stages, completed work, pending work, and key evidence, simplifying technical information and highlighting the case's progress.

[0052] Operation button control: Operation buttons are dynamically loaded based on user permissions for information objects. Buttons for which the user lacks permission are not displayed to prevent accidental user actions. For example: Users with editing privileges will see the "Add Comment" button displayed. Users with submission privileges will see a "Submit to Court, Send to the Other Party" button displayed. Users with only viewing permissions will only see the "View Download" button (if they have download permissions).

[0053] Output a personalized, dynamic view that includes a list of information that the user has the right to view, the fields displayed for the information, the relationships between the information, and the action buttons that the user can perform. This view is rendered in real time on the user's access interface.

[0054] The process status synchronization and triggering module is configured to synchronize the actual process stages of the case and publish global events, triggering batch re-evaluation of policy rules and driving the view to change dynamically with the process. Specifically, it includes: This module ensures that the system's phase dimensions are synchronized with the actual legal process in real time, and is the core driving force behind the dynamic changes in the view.

[0055] Its core value lies in triggering the batch application of strategy rules through automatic updates at each stage, thereby enabling the view to automatically adjust as the case progresses.

[0056] Programmatic node monitoring submodule: Function: Through a dual approach of system integration and manual confirmation, it accurately captures changes in procedural nodes of a case and updates the official stage of the case.

[0057] System integration mechanism: It integrates with external judicial systems such as the court's trial process management system and the arbitration institution's case management system, and uses REST API and message queues to achieve real-time synchronization. When external judicial systems generate procedural changes (such as issuing evidence exchange orders or court hearing announcements), they push event notifications to this system via REST API. This system receives notifications through a message queue to ensure that notifications are not lost or duplicated; The system automatically parses the notification content, extracts information such as case number, new stage name, and node trigger time, and updates the official stage of the case; Manual confirmation mechanism: For cases that cannot be connected to external judicial systems (such as pre-litigation mediation cases and foreign-related cases), a manual confirmation process is provided. The case administrator (or the lead attorney) submits a stage change application in the system based on the actual progress of the case, filling in information such as the new stage name, reason for change, and trigger time; After the application is submitted, the system will automatically send an approval notification to the relevant reviewer (such as the lead lawyer or the client's authorized agent). Once the reviewer approves the application, the official stage of the case is updated; if the reviewer fails to approve the application, the application is rejected, but the stage remains unchanged.

[0058] After a case is updated, the system automatically pushes a stage change notification to all parties involved in the case. The notification includes the case name, original stage, new stage, change time, and reason for change, ensuring that all parties are informed of the case's progress in a timely manner.

[0059] After the phase update operation is completed, the system publishes the operation as a global event, triggering the three-dimensional dynamic strategy matrix management module to re-evaluate the execution permissions; The latest official stage of the case is finally output and stored in the case status database, while a global event for stage change is also published. Policy rule batch activation submodule: Function: During the monitoring phase, global events are changed. Once the case phase changes, the permissions of all information objects in the case are automatically reassessed for all roles to ensure that the permission rules take effect synchronously with the case phase.

[0060] Real-time monitoring of process status synchronization and global events for stage changes published by the triggering module; when an event is captured, automatic permission re-evaluation of the process is initiated. The assessment scope supports two modes: full assessment and incremental assessment. Full assessment: A reassessment of the permissions of all information subjects in this case, applicable to major changes at each stage (such as moving from the first instance stage to the second instance stage). Incremental assessment: Only information objects related to the new phase (such as information whose target phase is the new phase) are subject to a reassessment of permissions. This is applicable to phase fine-tuning (such as moving from a pre-trial conference to a trial investigation).

[0061] The system defaults to incremental evaluation to improve evaluation efficiency, but users can manually select full evaluation.

[0062] Based on the new case stage, the policy rule base of the three-dimensional dynamic policy matrix management module is invoked to recalculate the permission set for each role for information objects within the selected range. The reassessed permission results are updated to the information permission database, overwriting the original permission records, and a permission change log is recorded (including information object ID, role, original permission, new permission, change time, and trigger rule ID). After the permission assessment is completed, the system automatically pushes a permission change notification to the roles whose permissions have changed. The notification includes the name of the information object, the content of the permission change, and the reason for the change (phase change), ensuring that users are aware of their permission changes in a timely manner.

[0063] Example: When a case progresses from the evidence exchange stage to the trial stage, the system automatically performs the following permission adjustments: All permissions for the judge's role regarding the exchanged evidence have been updated from pending to viewable, downloadable, and able to view annotations; All trial outlines have been changed from being editable and viewable to being read-only and viewable for the lawyers on their own side. The access rights of the opposing counsel to all cross-examination opinions have been updated from no access to view to view and download.

[0064] Output the updated set of information permissions and store it in the information permission database to ensure that the permissions queried by the view composition engine are the latest valid permissions.

[0065] The audit and strategy traceability module is configured to record full-dimensional, tamper-proof operation logs, providing simulation analysis of the impact before strategy modifications, ensuring the compliance of system operations and the rationality of strategies; Specifically, it includes: This module provides end-to-end trust assurance, meeting the stringent requirements of legal work for complete process traceability and auditability. Its core value lies in ensuring the compliance of system operations and the rationality of policy rules through panoramic operation logs and policy impact analysis.

[0066] Panoramic Operation Log Submodule: Function: To record all operations in the system in a comprehensive and tamper-proof manner, not only recording who did what and when, but more importantly, recording the case stage of the operation and the policy rules on which the operation was based, so as to achieve full-context traceability of the operation.

[0067] The log uses a structured format, and each log entry contains the following core fields: [Log ID, timestamp, user ID, user role, case number, case stage, operation type, information object ID, information object type, triggered policy rule ID, operation result, client IP, operation terminal]; The triggered policy rule ID field is used to associate the permission rule on which the operation is based. If there is no relevant policy rule for the operation (such as a system configuration operation), then this field is empty. The logs cover all core system operations, including: Information operations: viewing, editing, commenting, downloading, linking, printing, sharing, submitting, etc.; Rule operations: adding, modifying, deleting, enabling, and disabling policy rules, etc.; Phase-by-phase operations: submitting phase change requests, approving phase change requests, updating case phases, etc.; View operations: switching working intents, generating relationship maps, exporting views, etc.; Blockchain technology and timestamp signatures are used to ensure that the logs are immutable. Each log entry is automatically generated with a unique timestamp signature. Logs are stored in batches on blockchain nodes, with each node holding a complete copy of the logs. Any modification made to the logs by any node will be unverifiable by other nodes, ensuring the integrity and authenticity of the logs. It offers multi-dimensional log query functionality, supporting combined queries based on timestamps, user roles, case stages, operation types, policy rule IDs, and other conditions, facilitating auditors to quickly locate relevant operations. Furthermore, it supports exporting query results as audit reports in PDF and Excel formats to meet compliance review requirements. Output a complete, tamper-proof, and traceable panoramic operation log, stored in a blockchain log database, supporting querying, exporting, and auditing.

[0068] Strategy Impact Analysis and Simulator Submodule: Function: Before modifying policy rules (adding, modifying, deleting), it provides a policy impact simulation analysis function to simulate the impact of new rules on historical or current case information views, predict changes in the scope of information exposure, assist users in making reasonable rule adjustment decisions, and avoid sensitive information leakage or insufficient information access due to rule modification.

[0069] Simulation scenario selection: Supports two simulation scenarios: Historical Stage Simulation: Select a historical stage of the case and simulate the impact of the new rules on the information view at that stage; Future Stage Prediction: Select a possible future stage of the case (such as the second instance stage) and simulate the impact of the new rules on the information view at that stage; The system displays a comparison of the logical expressions of the original and new rules, clearly presenting the modifications made to the rules (such as expanding the scope of roles and increasing the set of permissions). Impact Scope Analysis: Simulating the changes in permissions of information objects after the new rules take effect, and compiling statistics: The number and type of information objects added to the permissions; The number and types of information objects whose permissions are reduced; Changes in the scope of information exposure (e.g., which new roles can view certain types of information); Visualization techniques such as bar charts, pie charts, and comparison tables are used to display the simulation analysis results, intuitively presenting the scope and extent of the new rules' impact. For example, a comparison table shows the changes in permissions for different roles under the old and new rules for a certain type of information; a bar chart shows the distribution of the number of information objects whose permissions have increased / decreased.

[0070] It supports generating simulation reports from simulation analysis results. The reports include simulation scenarios, rule comparisons, impact scope analysis, visualization charts, and risk warnings (such as the possibility that adding new permissions may lead to the leakage of sensitive information), making it convenient for users to submit for approval or archive. Before modifying core policy rules (such as adjusting access roles for absolutely confidential information), users can use this submodule to conduct simulation analysis to assess the risks of rule modification. If the risks are too high, users can abandon the modification or adjustment of the rule content to ensure the security and rationality of the policy rules.

[0071] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0072] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0073] It should be noted that, in this document, the use of relational terms such as "first" and "second" is merely for distinguishing one entity or operation from another, and does not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "include," "contain," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the statement "includes a…" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0074] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0075] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0076] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0077] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0078] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0079] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0080] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A smart legal collaborative processing system based on legal knowledge graphs, characterized in that: include: The legal information ontology and spatiotemporal label module is configured to standardize the type definition and assign contextual spatiotemporal labels to legal collaborative information, laying the data foundation for the generation of dynamic views; The three-dimensional dynamic strategy matrix management module is configured as a dynamic rule engine to build relationships between roles, stages, and information attributes, enabling the definition, reasoning, and stage-based adaptation of permissions. The real-time view composition module is configured to capture a three-dimensional context of user role, case stage, and current intent, and assemble a personalized interface that meets permissions and requirements in real time. The process status synchronization and triggering module is configured to synchronize the actual process stages of the case and publish global events, triggering batch re-evaluation of policy rules and driving the view to change dynamically with the process. The audit and strategy traceability module is configured to record full-dimensional, tamper-proof operation logs, providing simulation analysis of the impact before strategy modifications, ensuring the compliance of system operations and the rationality of strategies.

2. The intelligent legal collaborative processing system based on legal knowledge graph as described in claim 1, characterized in that, The legal information ontology and spatiotemporal tagging module specifically includes: Define the types and attribute relationships of all information objects in legal collaboration to form a dynamically extensible lightweight ontology; The system adopts a hierarchical classification structure, with the top level divided into four major categories: litigation documents, evidence materials, internal collaborative information, and notices from judicial organs. Each major category is further subdivided into specific types. The semantic relationships and constraints between information objects are defined, and are divided into three core relationships: support relationships, adversarial relationships, and evolutionary relationships. Standardized type labels are assigned to every document and every piece of data within the system, and these labels contain hierarchical type paths. Attach dynamically adjustable contextual tags to each piece of information; Four-dimensional structured tags are used: Generation phase: Specific litigation process nodes where marker information is generated; Target Phase: The specific phase in which information should be disclosed or used according to law or policy, and the associated triggering conditions; Strategic security classification: Based on the sensitivity of the information and the scope of use, it is divided into four levels, including public, public to the court, confidential to our party, and absolutely confidential; Information granularity: The scope of application of tagging, divided into document level, paragraph level, and field level; Tag assignment supports batch templated and personalized adjustments; Each output message carries a structured label: [type, generation stage, target stage, policy security level, information granularity].

3. The intelligent legal collaborative processing system based on legal knowledge graph as described in claim 1, characterized in that, The three-dimensional dynamic strategy matrix management module specifically includes: Define the dynamic permission mapping relationship between Role, Stage, and Info attribute; The core of the model is the rule function: Permission=F(Role,Stage,Info_Type,Info_Generate_Stage,Info_Target_Stage,Info_Security_Level,Info_Granularity); In this context, Role corresponds to a character role; Stage corresponds to a phase. The Info attribute contains: Information type (Info_Type), information generation stage (Info_Generate_Stage), information target stage (Info_Target_Stage), information security level (Info_Security_Level), and information granularity (Info_Granularity); The function outputs a set of operation permissions; Role: A hierarchical role system is adopted, divided into core roles and sub-roles: Stage: Consistent with the generation stage and target stage of the legal information ontology and spatiotemporal tagging module; Information Attribute Info: The information ontology tags and spatiotemporal tags that integrate the legal information ontology and spatiotemporal tag module include information type, generation stage, target stage, policy security level, and information granularity; Define rule priorities; when multiple rules conflict, the highest-priority rule will be executed. The final output is a policy rule base containing all the rules.

4. The intelligent legal collaborative processing system based on legal knowledge graph as described in claim 3, characterized in that, It also includes a visual policy configurator submodule: Provides preset rule templates that cover the core rules for common case types; When configuring rules, rule conflicts are detected in real time, and the currently configured rules are compared with the existing rules in the rule base. The addition, modification, and deletion of rules shall take effect only after approval. Approved and effective policy rules are automatically synchronized to the policy rule library and become the rule basis for the view composition engine.

5. The intelligent legal collaborative processing system based on legal knowledge graph as described in claim 1, characterized in that, The real-time view composition module specifically includes: Input information collection: User Identity Role: Obtained through the system's identity authentication module, it includes the user's core role and sub-roles, and supports multiple roles overlapping; The current official stage of the case: obtained from the process status synchronization and triggering module, is the latest official stage of the case. Only the current stage is used as the core parameter for view composition. The current user intent is inferred from the user's actions, or the user can manually select the current task intent in the system interface. Contextual validation: Verify whether the user's role is a participant in the case; Verify whether the current stage of the case is a valid stage; Verify whether the user's intent matches the current stage; If the verification fails, prompt the user and adjust the context parameters; The final output is a triple (R, S, Intent), where R is the user's set of roles, S is the current case stage, and Intent is the user's current intent.

6. The intelligent legal collaborative processing system based on legal knowledge graph as described in claim 5, characterized in that, Also includes: Using a multi-condition query, the logical expression is to find all information objects that meet the following conditions: The current user role, within the current stage, has viewing or higher permissions for this information object; The target stage of this information object is less than or equal to the current stage, or the generation stage is highly relevant to the current user intent. The information granularity of this information object conforms to the current user's permission scope; The information object was not marked as invalid; After the query results are returned, the system assembles the view from three dimensions based on the permission rules and information attributes, including display field control, relationship display, and operation button control; Output a dynamic view that includes a list of information that the user has the right to view, the fields displayed for the information, the relationships between the information, and the action buttons that the user can perform.

7. The intelligent legal collaborative processing system based on legal knowledge graph as described in claim 1, characterized in that, The process status synchronization and triggering module specifically includes: It interfaces with external judicial systems and synchronizes in real time. For cases that cannot be integrated with external judicial systems, a manual confirmation process is provided: After a case is updated, a notification of the phase change will be sent to all parties involved in the case. After the phase update operation is completed, the operation is published as a global event for re-evaluation. The latest official stage of the case has been finalized. Real-time monitoring of process status synchronization and global events for stage changes published by the triggering module; when an event is captured, automatic permission re-evaluation of the process is initiated. Based on the new case stage, the policy rule base is invoked to recalculate the permission set for each role for the selected information objects. The re-evaluated permission results are updated in the information permission database, and the updated information permission set is output and stored in the information permission database.

8. The intelligent legal collaborative processing system based on legal knowledge graph according to claim 1, characterized in that, The audit and strategy traceability module specifically includes: The log uses a structured format, and each log entry contains the following core fields: [Log ID, timestamp, user ID, user role, case number, case stage, operation type, information object ID, information object type, triggered policy rule ID, operation result, client IP, operation terminal]; The triggered policy rule ID field is used to associate the permission rule on which the operation is based; The log records cover all core system operations, including information operations, rule operations, stage operations, and view operations.

9. The intelligent legal collaborative processing system based on legal knowledge graph as described in claim 8, characterized in that, Also includes: The simulation scenario selection supports historical stage simulation and future stage prediction; Display a comparison of the logical expressions of the original and new rules, and present the changes to the rules; Simulate the changes in information object permissions after the new rules take effect, and statistically analyze the number and types of information objects with increased permissions, the number and types of information objects with decreased permissions, and the changes in the scope of information exposure.

Images