Case material structured analysis and responsibility path generation system

Abstract

The application discloses a case material structured analysis and responsibility path generation system, and belongs to the technical field of digital governance and process automation. The system comprises a case material collection and structured input module, an abnormal behavior identification and rule matching module, a material credibility pre-classification and evidence chain logical analysis module, a relationship network construction and suspicious association marking module, and a classification report generation and responsibility path output module. The system automatically identifies abnormal behaviors and generates abnormal behavior tags by structurally processing case materials, combining a preset abnormal behavior mode library and a rule library; constructs a relationship network graph based on available relationship clues, generates an association degree mark; and generates a classification report according to evidence chain analysis and the relationship network graph, and can selectively output an evidence chain credibility report, a relationship network visualization report, a logical contradiction warning report, a responsibility attribution reference report, a follow-up action suggestion report and the like. The application is applicable to pre-litigation, litigation and post-litigation stages, and has cross-field universality.

Description

Technical Field

[0001] This invention relates to the field of digital governance and process automation technology. Specifically, it relates to a method and system for structured analysis of case materials, construction of relationship networks, and generation of responsibility paths, which can be applied to pre-litigation, during-litigation, and post-litigation stages. Background Technology

[0002] In the fields of judiciary, administration, healthcare, finance, and public works, the submission and analysis of case materials are fundamental steps in initiating and advancing proceedings. However, the existing system has shortcomings in the following aspects: First, case materials are presented in diverse formats, lacking a unified intelligent processing interface. Evidence in various formats, such as text, images, audio, and video, is difficult to automatically identify, extract, and analyze within a single system. Second, the logical relationships between pieces of evidence have not been systematically identified, resulting in many cases having broken chains of evidence or logical contradictions before entering formal proceedings, but these are difficult to detect in a timely manner. At the same time, the existing system is also unable to provide a self-consistency assessment report before case materials enter formal proceedings, making it difficult for the submitting party to understand the weaknesses in their own materials in a timely manner. Third, the identification of abnormal behavior relies on manual judgment and lacks a systematic rule comparison and pattern recognition mechanism, making it difficult to detect hidden procedural violations and abuses of power. Fourth, the relationships between responsible parties are not systematically recorded and utilized, resulting in many violations with inherent logical connections being treated as isolated incidents, failing to form an effective accountability chain. Fifth, the generation of accountability recommendations relies heavily on experience and judgment, lacking standardized generation paths and traceable generation logic, which leads to difficulties in initiating accountability and inaccurate accountability. Summary of the Invention

[0003] This invention provides a system for structured analysis of case materials and generation of responsibility paths, comprising the following steps: S1: Case Material Collection and Structured Input The system receives case materials uploaded by the submitter, classifies them into formats such as text, images, audio, and video, and extracts key entity information (subject, time, relationship, claim, and type of evidence). S2: Abnormal Behavior Identification and Rule Matching The system, based on a pre-built library of abnormal behavior patterns and rules, compares extracted descriptions of abnormal behaviors with known abnormal patterns, identifies behaviors that match the abnormal patterns, and generates abnormal behavior tags. The abnormal behavior pattern library and rule library are derived from laws, regulations, rules, industry standards, and internal management specifications of the organization. S3: Material Credibility Pre-classification and Evidence Chain Logical Analysis The system pre-classifies the credibility of materials (Level A: High credibility; Level B: Medium credibility; Level C: To be verified). The system performs logical consistency analysis on structured information, identifying supporting relationships between pieces of evidence, timeline consistency, and continuity of behavioral chains, and marking logical breakpoints and contradictions. S4: Relationship Network Construction and Suspicious Association Marking The system extracts information about the parties involved in the case, constructs a relationship network diagram, and automatically marks various association patterns such as address, occupation, relatives, financial transactions, and joint contacts, generating a relationship suspicion score. S5: Classification Report Generation and Responsibility Path Output Based on evidence chain analysis, relationship network diagrams, and rule matching results, the system generates classification reports and can selectively output one or more of the following report types: evidence chain credibility report, relationship network visualization report, logical contradiction warning report, responsibility attribution reference report, and follow-up action suggestion report. Beneficial effects Compared with the prior art, the present invention has the following beneficial effects: First, it lowers the barrier to processing case materials. Submitting parties do not need legal or technical expertise; they only need to upload materials according to the system prompts to obtain a structured analysis report. Second, it improves the efficiency and coverage of evidence chain analysis. The system can simultaneously process case materials in multiple formats, including text, images, audio, and video, and can automatically identify logical contradictions and breakpoints, effectively avoiding oversights that may occur when relying on manual review. Third, it enhances the ability to identify relationship networks. Based on identifiable clues, the system can automatically construct and visualize relationship network diagrams between the main parties involved in a case, providing investigators and submitters with more intuitive analytical references. Fourth, it enables the pre-generation of liability pathways. The system can generate preliminary liability pathways and risk warnings before formal litigation, whistleblowing, or review procedures are initiated, helping submitters and case handlers identify potential problems earlier. Fifth, it has cross-domain versatility. The system adopts a modular architecture design, which can be connected to rule bases and exception pattern bases of different fields, and is applicable to multiple fields such as justice, administration, medical care, finance, and public works. Attached Figure Description Figure 1 This is a flowchart of the system for structured analysis of case materials and generation of liability paths according to the present invention. Detailed Implementation To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are only for explaining the invention and are not intended to limit the scope of protection of the invention. Example 1: Generation of Responsibility Paths in the Administrative / Judicial Fields This embodiment takes the scenario of an abnormal administrative action causing the procedure to stagnate as an example. S1: The submitter submits case materials through the system portal, including records of administrative actions, relevant evidence documents, and explanations of relationships, pointing out that a certain administrative agency has engaged in procedural violations in a specific event. S2: The system compares the case materials with a pre-set abnormal behavior pattern library, automatically identifies abnormal patterns such as "program nodes are bypassed", "statutory time limit is exceeded", and "electronic documents are abnormally modified", and generates corresponding abnormal behavior tags. S3: Based on the relationship description provided by the submitter and the pre-stored personnel relationship database in the system, the system automatically constructs a relationship network diagram related to the abnormal behavior and identifies the levels of responsible entities that may be involved. S4: The system matches abnormal behavior tags with a pre-set rule base to identify the type of violation, and identifies potentially responsible positions based on the relationship network diagram, automatically generating a responsibility path. S5: The system outputs a preliminary responsibility attribution reference report, including a description of the abnormal behavior, the type of violation, a suggestion for identifying the responsible person, a suggested responsibility path, and prompts for subsequent processing steps, for the submitter and relevant processing agencies to refer to. Example 2: Generation of Responsibility Paths in Public Funding / Engineering Sector This example illustrates a scenario where irregularities in the use of public funds lead to the loss of public funds. S1: The submitter submits case materials through the system portal, including fund flow records, project approval documents and related personnel information, pointing out that there are procedural irregularities or abnormal approvals in a certain public fund usage process. S2: The system compares the case materials with a pre-set abnormal behavior pattern library, automatically identifies abnormal patterns such as "mismatch between fund disbursement node and approval node", "missing approval procedure", and "inconsistency between acceptance record and payment record", and generates corresponding abnormal behavior tags. S3: Based on the relationship descriptions provided by the submitter and the pre-stored personnel relationship database, the system constructs a network diagram of the responsible persons related to the use of funds, and identifies the possible chains of responsibility involved. S4: The system matches abnormal behavior tags with a pre-set rule base to identify the approvers, appropriation reviewers, and superior supervisors who may be responsible, and automatically generates a responsibility path. S5: The system outputs a preliminary responsibility attribution reference report, including the path of fund violations, suggestions for identifying responsible persons, suggestions for responsibility paths, and prompts for subsequent handling steps, for the submitter and relevant supervisory agencies to refer to. Example 3: Preliminary Analysis of Administrative Litigation Materials In this embodiment, before formally filing an administrative lawsuit, the submitting party (plaintiff) uploads the preliminary case materials (statement, screenshots of evidence, timeline) to the system. S1: The system performs format recognition and key information extraction on the uploaded materials, extracting the subject, time point, main demands and existing evidence. S2: The system identified some evidence as having timestamps and third-party sources (hospital records, police reports), and pre-classified it as Grade A; other materials only have statements from the parties involved, and are tentatively classified as Grade B. S3: The system logically splices together the timeline and evidence, and finds two inconsistencies between the time points and the event descriptions, which are marked as "logical warning points" in the report. S4: The system extracts the main information of the two parties involved in the case and marks them as "potential connections" based on available relationship clues (address, occupation, relatives, financial transactions, mutual contacts, etc.). S5: The system outputs a structured analysis report, including an evidence chain credibility score, a relationship network diagram, logical warning annotations, preliminary suggestions for liability attribution, and recommendations for follow-up actions. The submitting party supplemented the missing evidence based on the report and submitted it to the court. Example 4: Relationship Network Analysis of Civil Appeal Materials In this embodiment, when preparing the appeal petition, the submitter (appellant) uploads the first-instance materials, new evidence, and relationship description to the system. S1: The system identifies the appeal petition, the first-instance judgment, new evidence documents (including audio), and relationship descriptions. S2: The system identifies new evidence containing audio files, calls the audio transcription module to transcribe the content, and incorporates it into the structured information database. S3: The system compares the grounds for appeal with the evidence from the first instance and finds that a fact pointed to by the grounds for appeal has no corresponding record in the evidence from the first instance, and marks it as "missing evidence warning". S4: The system extracts information on the appellant, appellee, and key witnesses. The system extracts information on the parties involved in the case, constructs a relationship network diagram based on available relationship clues (such as address, occupation, kinship, financial transactions, and joint contacts), and generates correlation markers accordingly. S5: The system outputs a structured analysis report, including a credibility score for the chain of evidence, a relationship network diagram, and an assessment of the logical consistency of the appeal grounds. After the appellant supplements the evidence based on the missing evidence warnings in the report, the appeal is formally submitted. Example 5: Logical Consistency Analysis of Medical Malpractice Identification Materials In this embodiment, the submitter (patient's family member) uploads medical records, imaging data, and family member statements to the system before applying for a medical malpractice assessment. S1: The system classifies medical records and imaging data, and extracts key time points, diagnosis and treatment behaviors, and records of changes in patient status. S2: The system identified the medical records and imaging data as being issued by the hospital and belonging to a traceable third-party source, and pre-classified as Grade A; the family's statement is tentatively classified as Grade B. S3: The system compares the family's statement with the medical record and finds a discrepancy between the time of the family's statement and the time recorded in the medical record, marking it as a "logic break warning". S4: The system extracts information about the parties involved in the case, constructs a relationship network diagram based on available relationship clues (address, occupation, relatives, financial transactions, co-contacts, etc.), and generates a correlation label accordingly. S5: The system outputs a structured analysis report, including a credibility score for the chain of evidence, warnings of logical breaks, and preliminary suggestions for liability attribution. Based on the warnings in the report, the submitting party supplemented relevant explanations and records before applying for expert evaluation.

Claims

1. A method for structured analysis of case materials and generation of responsibility paths, characterized in that, Includes the following steps: S1. Case Material Collection and Structured Input Steps: The system receives the case materials uploaded by the submitter, performs format recognition and classification on the materials, and extracts key entity information; S2. Abnormal behavior identification and rule matching steps: Based on the pre-set abnormal behavior pattern library and rule library, the extracted abnormal behavior descriptions are compared with known abnormal patterns to identify behaviors that match the abnormal patterns and generate abnormal behavior labels. S3. Material credibility pre-grading and evidence chain logic analysis steps: The system pre-grades the credibility of the materials and performs logical self-consistency analysis on the structured information, identifies the supporting relationships, timeline consistency and continuity of the behavioral chain between the evidence, and marks the logical breakpoints and contradictions. S4. Relationship Network Construction and Suspicious Association Marking Steps: The system extracts information on the parties involved in the case, constructs a relationship network diagram based on available relationship clues, and generates association degree markers or relationship suspicion degree scores; S5. Classification Report Generation and Responsibility Path Output Steps: Based on evidence chain analysis, relationship network diagrams, and rule matching results, the system generates classification reports and can selectively output one or more report types.

2. The method according to claim 1, characterized in that, In step S2, the abnormal behavior pattern library and rule library are derived from laws, regulations, rules and regulations, industry standards and internal management norms of the organization.

3. The method according to claim 1 or 2, characterized in that, In step S1, the key entity information includes subject information, time information, relationship information, claim information, and evidence type information.

4. The method according to claim 1, characterized in that, In step S3, the credibility pre-grading includes three levels: A, B, and C.

5. The method according to claim 1, characterized in that, In step S5, the report type includes one or more of the following: evidence chain credibility report, relationship network visualization report, logical contradiction warning report, responsibility attribution reference report, and follow-up action suggestion report.

6. A system for structured analysis of case materials and generation of responsibility paths, characterized in that, include: The case material collection and structured input module is used to perform step S1 as described in claim 1; An abnormal behavior identification and rule matching module is used to perform step S2 as described in claim 1; The material credibility pre-grading and evidence chain logic analysis module is used to perform step S3 as described in claim 1; A relational network construction and suspicious association marking module is used to perform step S4 as described in claim 1; The classification report generation and responsibility path output module is used to perform step S5 as described in claim 1.

Images