A legal compliance full life cycle management platform based on digital twinning

Abstract

This invention relates to the field of digital twin technology, and more particularly to a legal compliance full lifecycle management platform based on digital twins. By setting up a model building module, a unit invocation module, a behavior analysis module, an anomaly adjustment module, and a compliance verification module, a digital twin model is constructed and adjusted. Several behavioral data points are input into the model, and the behavioral data is analyzed to run the execution unit. The accuracy of the model's operation is judged based on the execution results. The anomaly adjustment module is invoked to determine the adjustment method based on abnormal operating states, and the adjusted model is subjected to compliance verification to determine its usability. The adjusted model is then put into use. This invention, by constructing a general digital twin model, performing targeted analysis and adjustment of single and complex behaviors, and determining the usability and specificity of the model, improves the accuracy of judgment and analysis efficiency of complex behaviors.

Description

Technical Field

[0001] This invention relates to the field of digital twin technology, and in particular to a legal compliance full lifecycle management platform based on digital twins. Background Technology

[0002] Faced with the severe challenges of a surge in global regulations, dynamic updates, high compliance costs, and soaring risks, as well as the common pain points of fragmented, static, and complex operational relationships in enterprise compliance management, the maturity of digital twin technology and its supporting technologies provides a key opportunity to build a legal compliance full lifecycle management platform. The core of this platform lies in using digital twins to construct a dynamic virtual mapping that integrates physical entities, business processes, and regulatory rules, thereby achieving a fundamental transformation in compliance risk management from passive response to proactive prediction and prevention, from static document management to dynamic monitoring of the entire process, and from information silos to unified intelligent collaboration, thus significantly improving the efficiency, accuracy, and foresight of compliance management.

[0003] Chinese Patent Publication No. CN118426922A discloses a model management method and system for a digital twin platform. The method includes integrating the model into a microservice architecture, debugging model parameters online, optimizing model return results, obtaining model interface calls through a REST interface, and formulating call strategies. It also analyzes the model's operational status and visualizes the model's usage details based on the call data. This invention provides a model management method for a digital twin platform that enables rapid model deployment and dynamic management, improves the flexibility and efficiency of model deployment, ensures the reliability and stability of model services, ensures high adaptability and accuracy of model output through meticulous parameter adjustments, and enhances the security and stability of model services by flexibly configuring usage permissions according to user and on-site needs. Through permission control and call strategies, this invention achieves better results in terms of flexibility, security, and stability.

[0004] Chinese Patent Publication No. CN118965230A discloses a smart operation and maintenance management platform based on digital twins. By confirming the storage capacity of image data models and text data models and selecting appropriate memory and sub-storage spaces for storage, efficient management of data models is achieved. The stored image data models and text data models are timestamped, giving these models time attributes and enhancing data traceability. By extracting feature data from the image and text data models separately and concatenating the feature vectors, a feature set containing rich information can be formed. According to the severity of the fault, corresponding early warning measures are taken to improve the pertinence and effectiveness of the early warning. Early warning alarms of different levels are transmitted to the abnormal area for alarm processing, ensuring that operation and maintenance personnel can understand and handle faults in a timely manner.

[0005] However, the following problems still exist in the existing technology.

[0006] In practice, there are complex behaviors. The legal provisions corresponding to the completion of a complex behavior are inconsistent with those corresponding to the incomplete complex behavior. Ignoring complex behaviors or misjudging the completion status of complex behaviors can lead to errors in the determination of legal provisions, thereby reducing the accuracy of model analysis and the compliance of legal provisions. Summary of the Invention

[0007] To address this issue, the present invention provides a digital twin-based legal compliance lifecycle management platform to resolve the problem that, in practice, there are complex behaviors, and the legal provisions corresponding to the completed complex behaviors are inconsistent with those corresponding to the incomplete complex behaviors. Ignoring complex behaviors or misjudging the completion status of complex behaviors can lead to errors in the determination of legal provisions, thereby reducing the accuracy of model analysis and the compliance of legal provisions.

[0008] To achieve the above objectives, this invention provides a legal compliance full lifecycle management platform based on digital twins, comprising:

[0009] The model building module is used to acquire the data required to build a digital twin model and to build the model based on the data. The operating units in the model include single-behavior units and composite-behavior units.

[0010] The unit calling module is connected to the model building module to input several behavioral data into the model and analyze each behavioral data to run the running unit;

[0011] The behavior analysis module is connected to the unit calling module. In response to the running status of the single behavior unit, it obtains the single behavior running rules and compares the single behavior running rules with the single behavior input rules to determine whether the running of the single behavior unit is abnormal.

[0012] In response to the operating state of the composite behavior unit, composite behavior operation rules are obtained. The composite behavior operation rules are compared with composite behavior input rules to obtain operating characteristic coefficients, so as to determine whether the operation of the composite behavior unit is abnormal.

[0013] An abnormal adjustment module, which is connected to the unit calling module and the behavior analysis module, determines the adjustment method of the model in response to abnormal operation of the running unit;

[0014] The compliance verification module, which is connected to the behavior analysis module and the anomaly adjustment module, is used to verify the model, determine the usability of the adjusted model, and put the adjusted model into use.

[0015] Furthermore, the unit invocation module analyzes each of the behaviors to run the execution unit, including,

[0016] Used to extract keywords from the behavioral data in order to determine the number of keywords;

[0017] If the number of keywords is 1, then a single-behavior unit will be invoked.

[0018] If the number of keywords is greater than one, then the composite behavior unit will be invoked.

[0019] Furthermore, the behavior analysis module compares the single-behavior execution rule with the single-behavior input rule, including,

[0020] Used to determine the correspondence between the single-line running rule and the single-line input rule;

[0021] The correspondence includes the fact that the content of the single-line running rule is consistent with the content of the single-line input rule.

[0022] Furthermore, the behavior analysis module determines whether the operation of the single behavior unit is abnormal, wherein,

[0023] If the correspondence is correct, then the single-behavior unit is determined to be operating normally.

[0024] If the correspondence is not found, then the operation of the single-behavior unit is determined to be abnormal.

[0025] Furthermore, the behavior analysis module obtains operational characteristic coefficients, including:

[0026] The number of corresponding rules is used to determine the correspondence between the compound behavior operation rules and the compound behavior input rules;

[0027] The ratio of the corresponding number of legal provisions to the number of legal provisions input by the composite behavior is used to determine the identity influence factor.

[0028] Used to determine the absolute value of the difference between the number of compound behavior operation rules and the number of compound behavior input rules;

[0029] The ratio of the number of composite behavior input method entries to the absolute value of the difference is used to determine the loss impact factor;

[0030] The weighted sum of the identity influence factor and the missing influence factor is used to determine the operating characteristic coefficient.

[0031] Furthermore, the behavior analysis module determines whether the operation of the composite behavior unit is abnormal, wherein,

[0032] If the operating characteristic coefficient is greater than the operating characteristic coefficient threshold, then the composite behavior unit is determined to be operating normally.

[0033] If the operating characteristic coefficient is less than or equal to the operating characteristic coefficient threshold, then the operation of the composite behavior unit is abnormal.

[0034] Furthermore, the anomaly adjustment module determines the adjustment method for the model, wherein,

[0035] If the operating unit meets the first operating condition, then the model is reconstructed;

[0036] If the operating unit meets the second operating condition, then the compound behavior operating rules are analyzed to determine the adjustment method of the model;

[0037] The first operating condition is that the single-behavior unit is malfunctioning, and the second operating condition is that the single-behavior unit is malfunctioning normally while the composite-behavior unit is malfunctioning.

[0038] Furthermore, the anomaly adjustment module analyzes the compound behavior operation rules, including,

[0039] Used to determine the associated keywords of the behavior data corresponding to the composite behavior unit;

[0040] This is used to determine the compound behavior operation rule corresponding to the associated keyword as the associated compound behavior operation rule;

[0041] This is used to determine that the composite behavior input method corresponding to the associated keyword is an associated composite behavior input method;

[0042] The ratio of the number of execution rules for the associated composite behavior to the number of input rules for the associated composite behavior is used to determine the abnormal behavior coefficient.

[0043] Furthermore, the anomaly adjustment module determines the adjustment method for the composite behavior unit, wherein,

[0044] If the abnormal behavior coefficient is greater than the abnormal behavior coefficient threshold, the breakage relationship of the associated keywords is determined, so as to perform a predetermined number of analyses on the behavior data corresponding to the absence of breakage.

[0045] If the abnormal behavior coefficient is less than or equal to the abnormal behavior coefficient threshold, then the single behavior unit is invoked.

[0046] Furthermore, the compliance verification module determines the usability of the adjusted model, including,

[0047] Used to input known composite behaviors into the adjusted model;

[0048] This is used to determine whether the content of the corresponding composite behavior input rule is consistent with the content of the composite behavior execution rule;

[0049] If the content remains consistent, then the adjusted model is deemed usable;

[0050] If the content is inconsistent, the adjusted model is deemed unusable.

[0051] Compared with existing technologies, this invention constructs and adjusts digital twin models by setting up a model building module, a unit invocation module, a behavior analysis module, an anomaly adjustment module, and a compliance verification module. Several behavioral data points are input into the model, and the behavioral data is analyzed to run the execution unit. The accuracy of the model's operation is judged based on the execution results. The anomaly adjustment module is invoked to determine the adjustment method based on abnormal operating states. The adjusted model is then subject to compliance verification to determine its usability before being put into use. This invention constructs a general digital twin model, performs targeted analysis and adjustment of single and complex behaviors, determines the usability and specificity of the model, and improves the accuracy and efficiency of judging complex behaviors.

[0052] In particular, this invention divides the model into single-behavioral units and composite-behavioral units to achieve refined analysis of behavioral data. In practice, the multiple sub-behaviors contained in behavioral data need to satisfy specific logical relationships. If the logical sequence of sub-behaviors is not coherent or does not reach a logically complete state, the behavior cannot be identified as a valid composite behavior. Incorrectly identifying single behaviors or incomplete logical sequences as composite behaviors not only risks the application of legal provisions but also leads to unnecessary consumption of computational resources, reducing the accuracy of model analysis and the compliance of legal provisions. Furthermore, existing legal compliance analysis models generally cannot effectively distinguish the legal logical differences between single behaviors and composite behaviors, resulting in incorrect legal conclusions even when the logical chain in a composite behavior is broken. Based on this, this invention considers targeted division of the model's operating units to achieve precise adaptation of behavioral granularity and improve analysis efficiency and accuracy.

[0053] In particular, this invention analyzes the operational status of composite behavior units and compares the operational rules of composite behavior with the input rules of composite behavior to determine the accuracy of the operation of composite behavior units. This provides a theoretical basis for subsequent adjustments to composite behavior units. In real-world applications, input behavior datasets often contain a large number of independent single behavior instances. Such data can be directly processed by single behavior units without going through a complex composite behavior analysis process, and can be efficiently and accurately mapped to the corresponding single rule. However, due to the inherent generalization error or boundary ambiguity of the underlying behavior recognition or classification model, there is a phenomenon of misclassifying or incorrectly assigning isolated behavior instances that should be processed by single behavior units to composite behavior units. This leads to biases in the analysis of behavior data, reduces analysis efficiency and accuracy, and ultimately outputs incorrect rules. Based on this, this invention analyzes the operational status of composite behavior units to determine their operational status, providing a theoretical basis for whether to adjust composite behavior units in the future, thereby improving the analysis efficiency and accuracy of the model.

[0054] In particular, this invention specifically determines whether to adjust single-behavioral units and composite-behavioral units. It is understood that single-behavioral units do not require multi-step analysis and can directly identify keywords for matching legal provisions. Meanwhile, the core feature of composite-behavioral units lies in the logical or temporal relationships between their internal constituent elements. The completeness of these relationships (i.e., all related behaviors have been implemented) and the accuracy of the implementation order are key prerequisites for determining whether the behavior as a whole meets the statutory constituent elements of a specific composite behavior. If the relationships are not fully realized or the implementation order is misaligned, the behavior as a whole loses its legal basis for constituting a composite behavior. In this case, to ensure the effectiveness of the model, it is necessary to analyze the data within the composite-behavioral unit. Data that can be analyzed by single-behavioral units can be directly transmitted to single-behavioral units, and the remaining data can be analyzed. This ensures the utilization rate of computing power and the reliability of model analysis, and improves the accuracy of legal fact determination and the reliability of the overall model output. Attached Figure Description

[0055] Figure 1 A schematic diagram of the structure of a digital twin-based legal compliance lifecycle management platform as an embodiment of the invention;

[0056] Figure 2 A logic block diagram for analyzing the described behaviors to run the execution unit according to an embodiment of the invention.

[0057] Figure 3 A logic block diagram for determining whether the operation of the single-behavior unit is abnormal, as shown in an embodiment of the invention;

[0058] Figure 4 A logic block diagram for determining whether the operation of the composite behavior unit is abnormal, as shown in an embodiment of the invention;

[0059] Figure 5 This is a logic block diagram of a method for determining the adjustment of the composite behavior unit according to an embodiment of the invention.

[0060] Figure 6 A logic block diagram for determining the availability of the adjusted model in an embodiment of the invention. Detailed Implementation

[0061] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0062] It should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the term "connection" should be interpreted broadly. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0063] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0064] Please see Figure 1 , Figure 1 This is a schematic diagram of the structure of a digital twin-based legal compliance lifecycle management platform according to an embodiment of the invention. The digital twin-based legal compliance lifecycle management platform of the present invention includes:

[0065] The model building module is used to acquire the data required to build a digital twin model and to build the model based on the data. The operating units in the model include single-behavior units and composite-behavior units.

[0066] The unit calling module is connected to the model building module to input several behavioral data into the model and analyze each behavioral data to run the running unit;

[0067] The behavior analysis module is connected to the unit calling module. In response to the running status of the single behavior unit, it obtains the single behavior running rules and compares the single behavior running rules with the single behavior input rules to determine whether the running of the single behavior unit is abnormal.

[0068] In response to the operating state of the composite behavior unit, composite behavior operation rules are obtained. The composite behavior operation rules are compared with composite behavior input rules to obtain operating characteristic coefficients, so as to determine whether the operation of the composite behavior unit is abnormal.

[0069] An abnormal adjustment module, which is connected to the unit calling module and the behavior analysis module, determines the adjustment method of the model in response to abnormal operation of the running unit;

[0070] The compliance verification module, which is connected to the behavior analysis module and the anomaly adjustment module, is used to verify the model, determine the usability of the adjusted model, and put the adjusted model into use.

[0071] Specifically, the model is updated periodically to ensure timely access to legal provisions. The update cycle is one quarter. Of course, those skilled in the art can determine the update cycle according to the actual situation, as long as it is reasonable, which will not be elaborated here.

[0072] Specifically, there are no restrictions on how the data and behavioral data required to build the digital twin model are obtained. For example, it can be open-source data published on the Internet or data provided by technical personnel, as long as authorization is obtained. This will not be elaborated further.

[0073] Specifically, single-line input law provisions are law provisions corresponding to a single line, with a one-to-one relationship between line and law provision; composite-line input law provisions are law provisions corresponding to composite lines, with a many-to-one relationship between line and law provision.

[0074] Specifically, this invention divides the model into single-behavioral units and composite-behavioral units to achieve refined analysis of behavioral data. In practice, the multiple sub-behaviors contained in behavioral data need to satisfy specific logical relationships. If the logical sequence of sub-behaviors is incoherent or does not reach a logically complete state, the behavior cannot be identified as a valid composite behavior. Incorrectly identifying single behaviors or incomplete logical sequences as composite behaviors not only risks the application of legal provisions but also leads to unnecessary consumption of computational resources, reducing the accuracy of model analysis and the compliance of legal provisions. Furthermore, existing legal compliance analysis models generally cannot effectively distinguish the legal logical differences between single behaviors and composite behaviors, resulting in incorrect legal conclusions even when the logical chain in a composite behavior is broken. Based on this, this invention considers targeted division of the model's operating units to achieve precise adaptation of behavioral granularity and improve analysis efficiency and accuracy.

[0075] Please see Figure 2 , Figure 2This is a logic block diagram illustrating the analysis of each of the described behaviors to run the execution unit, as per an embodiment of the invention. Specifically, the unit invocation module analyzes each of the described behaviors to run the execution unit, including:

[0076] Used to extract keywords from the behavioral data in order to determine the number of keywords;

[0077] If the number of keywords is 1, then a single-behavior unit will be invoked.

[0078] If the number of keywords is greater than one, then the composite behavior unit will be invoked.

[0079] Specifically, there are no restrictions on the method of keyword extraction. For example, the TF-IDF method can be used to initially screen candidate words, and then filtered through a legal terminology database. Combined with contextual analysis of legal elements (such as subjects, rights, obligations, and responsibilities), related words can be extracted. Of course, those skilled in the art can also determine the selection method according to the actual situation, as long as the keywords can be extracted. This will not be elaborated further.

[0080] It is understandable that a legal terminology database can be an open-source database or a database of terms provided by professionals, which will not be elaborated further.

[0081] Specifically, the behavior analysis module compares the single-behavior execution rule with the single-behavior input rule, including,

[0082] Used to determine the correspondence between the single-line running rule and the single-line input rule;

[0083] The correspondence includes the fact that the content of the single-line running rule is consistent with the content of the single-line input rule.

[0084] Specifically, consistency in legal provisions means that all the texts related to the legal provisions can be matched one-to-one.

[0085] Please see Figure 3 , Figure 3 This is a logic block diagram illustrating how to determine whether the operation of the single behavior unit is abnormal, according to an embodiment of the invention. Specifically, the behavior analysis module determines whether the operation of the single behavior unit is abnormal, wherein...

[0086] If the correspondence is correct, then the single-behavior unit is determined to be operating normally.

[0087] If the correspondence is not found, then the operation of the single-behavior unit is determined to be abnormal.

[0088] Specifically, the behavior analysis module obtains operational characteristic coefficients, including,

[0089] The number of corresponding rules is used to determine the correspondence between the compound behavior operation rules and the compound behavior input rules;

[0090] The ratio of the corresponding number of legal provisions to the number of legal provisions input by the composite behavior is used to determine the identity influence factor.

[0091] Used to determine the absolute value of the difference between the number of compound behavior operation rules and the number of compound behavior input rules;

[0092] The ratio of the number of composite behavior input method entries to the absolute value of the difference is used to determine the loss impact factor;

[0093] The weighted sum of the identity influence factor and the missing influence factor is used to determine the operating characteristic coefficient.

[0094] Specifically, the sum of the weighting coefficients of the identity impact factor and the missing impact factor is 1. When adjusting the weighting coefficients, considering that in practice, missing legal provisions indicate errors in the model's analysis and have a more significant impact on the model's accuracy, the weighting coefficient of the identity impact factor is set to 0.4, and the weighting coefficient of the missing impact factor is set to 0.6.

[0095] Please see Figure 4 , Figure 4 This is a logic block diagram illustrating how to determine whether the operation of the composite behavior unit is abnormal, according to an embodiment of the invention. Specifically, the behavior analysis module determines whether the operation of the composite behavior unit is abnormal, wherein...

[0096] If the operating characteristic coefficient is greater than the operating characteristic coefficient threshold, then the composite behavior unit is determined to be operating normally.

[0097] If the operating characteristic coefficient is less than or equal to the operating characteristic coefficient threshold, then the operation of the composite behavior unit is abnormal.

[0098] Specifically, the operating characteristic coefficient threshold characterizes a boundary of the composite behavioral unit analysis anomaly. For pre-calculated data, several operating characteristic coefficients corresponding to the composite behavioral unit analysis anomaly are obtained. The product of the average value of each operating characteristic coefficient and the accuracy coefficient is determined as the operating characteristic coefficient threshold. The accuracy coefficient is selected in the interval [1, 1.2]. In practice, in order to improve the accuracy of the model, the accuracy coefficient is determined to be 1.1.

[0099] Specifically, this invention analyzes the operational status of composite behavior units, compares the operational rules of composite behavior with the input rules of composite behavior, and determines the accuracy of the operation of composite behavior units. This provides a theoretical basis for subsequent adjustments to composite behavior units. In real-world applications, input behavior datasets often contain a large number of independent single behavior instances. Such data should be directly routed to single behavior units for processing, without needing a complex composite behavior analysis process, to efficiently and accurately map to the corresponding single rule. However, due to the inherent generalization error or boundary ambiguity of the underlying behavior recognition or classification model, there is a phenomenon of misclassifying or incorrectly assigning isolated behavior instances that should be processed by single behavior units to composite behavior units. This leads to biases in the analysis of behavior data, reduces analysis efficiency and accuracy, and ultimately outputs incorrect rules. Based on this, this invention analyzes the operational status of composite behavior units to determine their operational status, providing a theoretical basis for whether to adjust composite behavior units subsequently, thereby improving the model's analysis efficiency and accuracy.

[0100] Specifically, the anomaly adjustment module determines the adjustment method for the model, wherein,

[0101] If the operating unit meets the first operating condition, then the model is reconstructed;

[0102] If the operating unit meets the second operating condition, then the compound behavior operating rules are analyzed to determine the adjustment method of the model;

[0103] The first operating condition is that the single-behavior unit is malfunctioning, and the second operating condition is that the single-behavior unit is malfunctioning normally while the composite-behavior unit is malfunctioning.

[0104] Understandably, when analyzing behavioral data, if there is an abnormal operation of a single behavioral unit, there is no need to consider the operating status of other units; the model can be directly reconstructed.

[0105] Specifically, the anomaly adjustment module analyzes the compound behavior operation rules, including,

[0106] Used to determine the associated keywords of the behavior data corresponding to the composite behavior unit;

[0107] This is used to determine the compound behavior operation rule corresponding to the associated keyword as the associated compound behavior operation rule;

[0108] This is used to determine that the composite behavior input method corresponding to the associated keyword is an associated composite behavior input method;

[0109] The ratio of the number of execution rules for the associated composite behavior to the number of input rules for the associated composite behavior is used to determine the abnormal behavior coefficient.

[0110] Specifically, related keywords refer to two or more keywords present when a compound action occurs. These keywords correspond to only one legal provision, rather than corresponding to the same number of legal provisions as the number of keywords.

[0111] It is understandable that Example 1 exists.

[0112] Wang deliberately faked a fall (pretending to be injured) in front of a slow-moving truck. After getting up, he blocked the vehicle and threatened the driver by banging on the door with an iron bar: "If you don't give me 5,000 yuan, I'll smash the car and call the police to say you caused an accident and fled the scene!" The driver was forced to transfer the money.

[0113] The related keywords are "threat" and "transfer".

[0114] Please see Figure 5 , Figure 5 This is a logic block diagram illustrating the method for determining the adjustment of the composite behavior unit according to an embodiment of the invention. Specifically, the exception adjustment module determines the method for adjusting the composite behavior unit, wherein...

[0115] If the abnormal behavior coefficient is greater than the abnormal behavior coefficient threshold, the breakage relationship of the associated keywords is determined, so as to perform a predetermined number of analyses on the behavior data corresponding to the absence of breakage.

[0116] If the abnormal behavior coefficient is less than or equal to the abnormal behavior coefficient threshold, then the single behavior unit is invoked.

[0117] Specifically, the abnormal behavior coefficient threshold represents a boundary where compound behaviors exist. It is calculated in advance by acquiring several behavioral data with compound behaviors, determining several abnormal behavior coefficients, and determining the product of the mean of each abnormal behavior coefficient and the behavior accuracy coefficient as the abnormal behavior coefficient threshold. The behavior accuracy coefficient is selected in the interval [0.8, 0.98]. In order to improve the calculation accuracy, the behavior accuracy coefficient is determined to be 0.9 in the implementation.

[0118] Specifically, the predetermined number of times is two. Of course, those skilled in the art can also determine the number of times according to the actual situation, which will not be elaborated here.

[0119] Specifically, the method for determining the break relationship is to identify the relationship between the associated keywords and the final behavior.

[0120] If the final action is caused by related keywords, then the relationship is considered intact.

[0121] If the final action is not caused by related keywords, then the relationship is considered broken.

[0122] Understandably, in Example 1, if the "threat" directly leads to the "transfer of funds," the relationship is not broken; if the other party voluntarily gives the money after the threat, the relationship is broken.

[0123] Specifically, this invention specifically determines whether to adjust single-behavioral units and composite-behavioral units. It is understood that single-behavioral units do not require multi-step analysis and can directly identify keywords for legal provision matching. Meanwhile, the core characteristic of composite-behavioral units lies in the logical or temporal relationships between their internal constituent elements. The completeness of these relationships (i.e., all related actions have been implemented) and the accuracy of their implementation order are key prerequisites for determining whether the entire behavior meets the statutory constituent elements of a specific composite behavior. If the relationships are not fully realized or the implementation order is misaligned, the entire behavior loses its legal basis for constituting a composite behavior. In this case, to ensure the model's effectiveness, the data within the composite-behavioral unit needs to be analyzed. Data that can be analyzed by single-behavioral units can be directly transmitted to the single-behavioral units, and the remaining data can be analyzed. This ensures the utilization rate of computing power and the reliability of model analysis, improving the accuracy of legal fact determination and the overall reliability of the model's output.

[0124] Please see Figure 6 , Figure 6 This is a logic block diagram illustrating the determination of the adjusted model's availability according to an embodiment of the invention. Specifically, the compliance verification module determines the adjusted model's availability by including:

[0125] Used to input known composite behaviors into the adjusted model;

[0126] This is used to determine whether the content of the corresponding composite behavior input rule is consistent with the content of the composite behavior execution rule;

[0127] If the content remains consistent, then the adjusted model is deemed usable;

[0128] If the content is inconsistent, the adjusted model is deemed unusable.

[0129] It is understandable that Example 2 exists.

[0130] Example 2 behavioral data: Li hit pedestrian Zhang (who did not die) with his car at night. After finding Zhang seriously injured and groaning, Li, fearing to take responsibility, dragged Zhang to a wooded area in the suburbs and abandoned him, resulting in Zhang's death due to lack of timely medical treatment.

[0131] Example 2 corresponds to the following compound behavior input law: Article 232 of the Criminal Law (Intentional Homicide);

[0132] The adjusted model applies the legal provision Article 232 of the Criminal Law (Intentional Homicide) to the complex behavior in Example 2.

[0133] As can be seen, if the content of the composite behavior input rule is consistent with the content of the composite behavior execution rule, then the adjusted model is confirmed to be usable.

[0134] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0135] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A legal compliance full lifecycle management platform based on digital twins, characterized in that, include: The model building module is used to acquire the data required to build a digital twin model and to build the model based on the data. The operating units in the model include single-behavior units and composite-behavior units. The unit calling module is connected to the model building module to input several behavioral data into the model and analyze each behavioral data to run the running unit; The behavior analysis module is connected to the unit calling module. In response to the running status of the single behavior unit, it obtains the single behavior running rules and compares the single behavior running rules with the single behavior input rules to determine whether the running of the single behavior unit is abnormal. In response to the operating state of the composite behavior unit, composite behavior operation rules are obtained. The composite behavior operation rules are compared with composite behavior input rules to obtain operating characteristic coefficients, so as to determine whether the operation of the composite behavior unit is abnormal. An abnormal adjustment module, which is connected to the unit calling module and the behavior analysis module, determines the adjustment method of the model in response to abnormal operation of the running unit; The compliance verification module, which is connected to the behavior analysis module and the anomaly adjustment module, is used to verify the model, determine the usability of the adjusted model, and put the adjusted model into use. The unit calls the module to analyze each of the behaviors in order to run the running unit, including, Used to extract keywords from the behavioral data in order to determine the number of keywords; If the number of keywords is 1, then a single-behavior unit will be invoked. If the number of keywords is greater than one, then the composite behavior unit will be invoked. The anomaly adjustment module determines the adjustment method for the model, wherein, If the operating unit meets the first operating condition, then the model is reconstructed; If the operating unit meets the second operating condition, then the compound behavior operating rules are analyzed to determine the adjustment method of the model; The first operating condition is that the single-behavior unit is malfunctioning, and the second operating condition is that the single-behavior unit is malfunctioning normally while the composite-behavior unit is malfunctioning. The anomaly adjustment module analyzes the compound behavior operation rules, including, Used to determine the associated keywords of the behavior data corresponding to the composite behavior unit; This is used to determine the compound behavior operation rule corresponding to the associated keyword as the associated compound behavior operation rule; This is used to determine that the composite behavior input method corresponding to the associated keyword is an associated composite behavior input method; The ratio of the number of operation rules for the associated composite behavior to the number of input rules for the associated composite behavior is used to determine the abnormal behavior coefficient. The anomaly adjustment module determines the adjustment method for the composite behavior unit, wherein, If the abnormal behavior coefficient is greater than the abnormal behavior coefficient threshold, the breakage relationship of the associated keywords is determined, so as to perform a predetermined number of analyses on the behavior data corresponding to the absence of breakage. If the abnormal behavior coefficient is less than or equal to the abnormal behavior coefficient threshold, then the single behavior unit is invoked.

2. The legal compliance full lifecycle management platform based on digital twins as described in claim 1, characterized in that, The behavior analysis module compares the single-behavior execution rule with the single-behavior input rule, including: Used to determine the correspondence between the single-line running rule and the single-line input rule; The correspondence includes the fact that the content of the single-line running rule is consistent with the content of the single-line input rule.

3. The legal compliance full lifecycle management platform based on digital twins as described in claim 2, characterized in that, The behavior analysis module determines whether the operation of the single behavior unit is abnormal, wherein, If the correspondence is correct, then the single-behavior unit is determined to be operating normally. If the correspondence is not found, then the operation of the single-behavior unit is determined to be abnormal.

4. The legal compliance full lifecycle management platform based on digital twins according to claim 3, characterized in that, The behavior analysis module obtains operational characteristic coefficients, including: The number of corresponding rules is used to determine the correspondence between the compound behavior operation rules and the compound behavior input rules; The ratio of the corresponding number of legal provisions to the number of legal provisions input by the composite behavior is used to determine the identity influence factor. Used to determine the absolute value of the difference between the number of compound behavior operation rules and the number of compound behavior input rules; The ratio of the number of composite behavior input method entries to the absolute value of the difference is used to determine the loss impact factor; The weighted sum of the identity influence factor and the missing influence factor is used to determine the operating characteristic coefficient.

5. The legal compliance full lifecycle management platform based on digital twins according to claim 1, characterized in that, The behavior analysis module determines whether the operation of the composite behavior unit is abnormal, wherein, If the operating characteristic coefficient is greater than the operating characteristic coefficient threshold, then the composite behavior unit is determined to be operating normally. If the operating characteristic coefficient is less than or equal to the operating characteristic coefficient threshold, then the operation of the composite behavior unit is abnormal.

6. The legal compliance full lifecycle management platform based on digital twins according to claim 1, characterized in that, The compliance verification module determines the usability of the adjusted model, including: Used to input known composite behaviors into the adjusted model; This is used to determine whether the content of the corresponding composite behavior input rule is consistent with the content of the composite behavior execution rule; If the content remains consistent, then the adjusted model is deemed usable; If the content is inconsistent, the adjusted model is deemed unusable.

Images