Automatic jurisdiction rule matching method based on knowledge graph

Abstract

The invention discloses an automatic jurisdiction rule matching method based on a knowledge graph, and the method comprises the steps: inputting case information which comprises a case type, a mark amount, a residence of a party, and a cause of action; data preprocessing: standardizing the input case information; constructing a knowledge graph; carrying out dynamic weight calculation; carrying out multi-dimensional matching; and judging whether the matched jurisdiction courts have conflicts or not, if not, outputting the jurisdiction courts, and otherwise, solving the conflicts through a preset priority rule. According to the method, the knowledge graph is introduced to integrate legal provisions, cases and trial elements, so that structured expression of rules is realized; the priorities of different jurisdiction factors are quantified through a dynamic weight model, and the matching accuracy is improved; the method is suitable for intelligent determination of the jurisdiction in the case filing stage.

Description

Technical Field

[0001] This invention relates to the field of smart court technology, and in particular to an automatic matching method for jurisdiction rules based on knowledge graphs. Background Technology

[0002] Traditional methods rely on manual table lookups or simple rule matching, which are inefficient and prone to errors (such as errors in calculating the target amount leading to errors in level jurisdiction).

[0003] It cannot handle complex combinations of jurisdictional rules (such as exclusive jurisdiction taking precedence over territorial jurisdiction), and requires manual intervention in conflict decision-making.

[0004] CN110162539A discloses an intelligent case jurisdiction judgment system, including: an extraction module for extracting element information from case information; a rule judgment module for obtaining geographical information matching the element information and searching for jurisdictional information associated with the geographical information; and a push module for pushing the obtained jurisdictional information to the user. This application, through element analysis in case documents and the application of jurisdictional authority based on legal logic, not only achieves precise analysis of case elements and supplements the legal business elements of jurisdictional review, thus improving the micro-level elements of law and meeting the needs of intelligent jurisdiction judgment; but also automatically determines whether the currently handling court has jurisdiction, and other courts with jurisdiction, based on the case details. This patent can only determine whether jurisdiction exists, but cannot automatically match the court with jurisdiction. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides a knowledge graph-based automatic matching method for jurisdictional rules that integrates legal provisions, cases, and trial elements to achieve structured expression of rules; quantifies the priority of different jurisdictional factors through a dynamic weight model to improve matching accuracy; and is applicable to the intelligent determination of the court of jurisdiction at the case filing stage.

[0006] The objective of this invention is achieved through the following technical solutions.

[0007] An automatic matching method for jurisdiction rules based on knowledge graphs, comprising the following steps:

[0008] 1) Input case information, which includes case type, amount in dispute, domicile of the parties, and cause of action;

[0009] 2) Data preprocessing: Standardizing the input case information;

[0010] 3) Construct a knowledge graph;

[0011] 4) Perform dynamic weight calculation;

[0012] 5) Perform multi-dimensional matching;

[0013] 6) Determine if there is a conflict after matching the courts of jurisdiction. If there is no conflict, output the court of jurisdiction. Otherwise, resolve the conflict according to the preset priority rules.

[0014] The standardization of the input case information in step 2) includes: unifying the unit of the amount in dispute; parsing the geographical information to the administrative division code; and mapping the cause of action to the standard classification of the "Provisions on Causes of Action in Civil Cases".

[0015] The knowledge graph entity types in step 3) include: legal provisions, jurisdictional rules, and regional nodes; the relationship definitions include: applicable: legal provisions → jurisdictional rules, belonging to: regional nodes → superior regions, and priority: jurisdictional rules → weight values.

[0016] In step 4), the dynamic weight = basic weight mapping + target amount adjustment coefficient + regional level adjustment coefficient. Basic weight mapping: the mapping from rule type to basic weight. If the rule type does not exist, the basic weight is 0. Calculate the target amount adjustment coefficient: add 5 weight points for every 10 million. Calculate the regional level adjustment coefficient: add 10 for provincial level and 5 for municipal level.

[0017] Step 5) specifically includes:

[0018] 5.1) Exclusive Jurisdiction Filtering: If the case type matches the exclusive jurisdiction rule, return the result directly;

[0019] 5.2) Jurisdiction matching: Jurisdiction rules are matched according to the amount in dispute and the type of case, and the rule with the highest weight is selected.

[0020] 5.3) Territorial jurisdiction matching: Based on the administrative division code of the party's place of residence, match the nearest court with jurisdiction.

[0021] The priority rule preset in step 6) is exclusive jurisdiction > hierarchical jurisdiction > territorial jurisdiction.

[0022] Compared with the prior art, the advantages of this invention are:

[0023] (1) Dynamic weight model:

[0024] Jurisdiction rule weights are calculated in real time based on case attributes (such as the amount in dispute and the domicile of the parties) and high-priority rules are matched first (such as exclusive jurisdiction weight > hierarchical jurisdiction > territorial jurisdiction).

[0025] (2) Multi-dimensional matching algorithm:

[0026] It combines a three-tiered filtering mechanism: geographic matching (the place of residence of the parties), level matching (amount in dispute / case type), and specific matching (such as real estate disputes).

[0027] (3) Conflict resolution mechanism:

[0028] When multiple rules apply, the optimal solution is automatically selected based on preset priority rules (such as exclusive jurisdiction > hierarchical jurisdiction > territorial jurisdiction).

[0029] This invention has the following advantages: improved efficiency: jurisdiction matching is automatically completed during the case filing stage, reducing manual verification time by more than 50%.

[0030] Accuracy: Reduce the jurisdictional error rate through a dynamic weighting model (e.g., the error rate in calculating the target amount is <0.5%).

[0031] Scalability: Supports rapid configuration of new jurisdiction rules (such as jurisdiction rules for free trade zone courts). Attached Figure Description

[0032] Figure 1 This is a flowchart of the matching process of the present invention. Detailed Implementation

[0033] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0034] like Figure 1 As shown, an automatic matching method for jurisdiction rules based on knowledge graphs includes the following steps:

[0035] 1) Input case information, which includes case type, amount in dispute, domicile of the parties, and cause of action;

[0036] 2) Data preprocessing: Standardizing the input case information;

[0037] 3) Construct a knowledge graph;

[0038] 4) Perform dynamic weight calculation;

[0039] 5) Perform multi-dimensional matching;

[0040] 6) Determine if there is a conflict after matching the courts of jurisdiction. If there is no conflict, output the court of jurisdiction. Otherwise, resolve the conflict according to the preset priority rules.

[0041] The standardization of the input case information in step 2) includes: unifying the unit of the amount in dispute (e.g., 10,000 yuan → yuan), parsing the geographical information to the administrative division code (e.g., "Chaoyang District, Beijing" → 110105), and mapping the cause of action to the standard classification in the "Provisions on Causes of Action in Civil Cases" (e.g., "Private Lending Disputes" → Contract Disputes).

[0042] The knowledge graph entity types in step 3) include: legal provisions (such as Article 18 of the Civil Procedure Law, which governs different levels of jurisdiction), jurisdictional rules (such as "amounts exceeding 10 million are under the jurisdiction of the intermediate court"), and geographical nodes (such as provinces / cities / districts). The relationship definitions include: application (legal provisions → jurisdictional rules), belonging (geographical nodes → superior regions), and priority (jurisdictional rules → weight values).

[0043] In step 4), the dynamic weight = basic weight mapping + target amount adjustment coefficient + regional level adjustment coefficient. Basic weight mapping: the mapping from rule type to basic weight. If the rule type does not exist, the basic weight is 0. Calculate the target amount adjustment coefficient: add 5 weight points for every 10 million. Calculate the regional level adjustment coefficient: add 10 for provincial level and 5 for municipal level.

[0044] Step 5) specifically includes:

[0045] 5.1) Exclusive Jurisdiction Filtering: If the case type matches the exclusive jurisdiction rule (e.g., real estate dispute → court where the real estate is located), the result is returned directly;

[0046] 5.2) Jurisdiction matching: The jurisdiction rules are matched according to the amount in dispute and the case type (e.g., patent dispute → intermediate court), and the rule with the highest weight is selected.

[0047] 5.3) Territorial jurisdiction matching: Based on the administrative division code of the party's place of residence, match the court with the nearest jurisdiction (e.g., district courts take precedence over city courts).

[0048] The priority rule preset in step 6) is exclusive jurisdiction > hierarchical jurisdiction > territorial jurisdiction.

[0049] Example of conflict handling:

[0050] A contract dispute case meets both territorial jurisdiction (District A court) and hierarchical jurisdiction (intermediate court), so the intermediate court should be given priority.

[0051] Implementation Case:

[0052] 1. Enter the case:

[0053] Case type: Copyright infringement dispute

[0054] Target amount: 20 million yuan

[0055] Address of the party concerned: Pudong New Area, Shanghai

[0056] 2. Matching process:

[0057] (1) Exclusive jurisdiction: There are no exclusive jurisdiction rules for copyright infringement disputes.

[0058] (2) Jurisdiction by level: If the amount in dispute is greater than RMB 10 million, it is matched with the Shanghai Intellectual Property Court (weight 80+10=90).

[0059] (3) Territorial jurisdiction: Pudong New Area Court (weight 60+5=65).

[0060] (4) Conflict resolution: The jurisdictional level has higher weight, and the final match is the Shanghai Intellectual Property Court.

Claims

1. A method for automatic matching of jurisdiction rules based on knowledge graphs, characterized by the following steps: include: 1) Input case information, which includes case type, amount in dispute, domicile of the parties, and cause of action; 2) Data preprocessing: Standardizing the input case information; 3) Construct a knowledge graph; 4) Perform dynamic weight calculation; 5) Perform multi-dimensional matching; 6) Determine if there is a conflict after matching the courts of jurisdiction. If there is no conflict, output the court of jurisdiction. Otherwise, resolve the conflict according to the preset priority rules.

2. The method for automatic matching of jurisdiction rules based on knowledge graphs according to claim 1, characterized in that... The standardization of the input case information in step 2) includes: unifying the unit of the amount in dispute; parsing the geographical information to the administrative division code; and mapping the cause of action to the standard classification of the "Provisions on Causes of Action in Civil Cases".

3. The method for automatic matching of jurisdiction rules based on knowledge graphs according to claim 1, characterized in that... The knowledge graph entity types in step 3) include: legal provisions, jurisdictional rules, and regional nodes; the relationship definitions include: applicable: legal provisions → jurisdictional rules, belonging to: regional nodes → superior regions, and priority: jurisdictional rules → weight values.

4. The method for automatic matching of jurisdiction rules based on knowledge graphs according to claim 1, characterized in that... In step 4), the dynamic weight = basic weight mapping + target amount adjustment coefficient + regional level adjustment coefficient. Basic weight mapping: the mapping from rule type to basic weight. If the rule type does not exist, the basic weight is 0. Calculate the target amount adjustment coefficient: add 5 weight points for every 10 million. Calculate the regional level adjustment coefficient: add 10 for provincial level and 5 for municipal level.

5. The method for automatic matching of jurisdiction rules based on knowledge graphs according to claim 1, characterized in that... Step 5) specifically includes: 5.1) Exclusive Jurisdiction Filtering: If the case type matches the exclusive jurisdiction rule, return the result directly; 5.2) Jurisdiction matching: Jurisdiction rules are matched according to the amount in dispute and the type of case, and the rule with the highest weight is selected; 5.3) Territorial jurisdiction matching: Based on the administrative division code of the party's place of residence, match the nearest court with jurisdiction.

6. The method for automatic matching of jurisdiction rules based on knowledge graphs according to claim 1 or 5, characterized in that... The priority rule preset in step 6) is exclusive jurisdiction > hierarchical jurisdiction > territorial jurisdiction.

Images