Cross-border scheme generation method and device based on cross-border rules, equipment and medium

Abstract

The application relates to the technical field of cross-border scheme generation based on cross-border rules, and discloses a cross-border scheme generation method, device, equipment and medium based on cross-border rules, wherein the method comprises the following steps: acquiring business scene features and business parameters in real time and inputting the business scene features and the business parameters into an intelligent knowledge graph, setting a confidence evaluation and a standardization processing mechanism, efficiently screening and standardizing related cases, and formulating a cross-border scheme in combination with a target cross-border transaction simulator. The application has the beneficial technical effect that, when legal clauses conflict, dynamic coordination can be performed through intelligent logic analysis, compliance risks caused by rule conflicts are avoided, and the limitations of a traditional rule engine when facing rapidly changing regional trade agreements and multi-modal supervision requirements are effectively solved. The application has a significant technical advantage in improving cross-border trade compliance, flexibility and efficiency, and provides enterprises with more competitive trade strategies and operation management capabilities.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cross-border scheme generation based on cross-border rules, and in particular to a cross-border scheme generation method, device and equipment based on cross-border rules and a medium. BACKGROUND

[0002] In the process of digitalization of cross-border trade, the high-frequency clause changes and multi-modal supervision requirements of regional trade agreements have exceeded the capability boundary of traditional rule engines. However, the traditional rule engine has the problem of time effectiveness of static rule library. The current mainstream engine relies on manually maintained hard-coded rules and cannot be updated in real time when regional policies are implemented and updated. Moreover, when there is a conflict between legal clauses, for example, when an enterprise is subject to two legal clauses at the same time, the traditional rule engine cannot real-time coordinate the logical conflict of legal clauses. SUMMARY

[0003] Therefore, it is necessary to propose a cross-border scheme generation method, device and equipment based on cross-border rules and a medium for the existing cross-border scheme generation problem based on cross-border rules.

[0004] A cross-border scheme generation method based on cross-border rules, the method comprising:

[0005] obtaining business scene features and business parameters of a specified business;

[0006] inputting the business scene features into a preset knowledge graph to obtain corresponding legal clause features;

[0007] obtaining a plurality of related cases based on the legal clause features, and performing confidence evaluation on the related cases to obtain confidence scores of the related cases; wherein the related cases contain time stamps;

[0008] based on the confidence scores of the related cases, performing preset standardization processing on the related cases respectively to obtain standardized cases corresponding to the related cases;

[0009] training a preset transaction simulator based on the standardized cases to obtain a target cross-border transaction simulator;

[0010] inputting the business parameters into the target cross-border transaction simulator to obtain a cross-border scheme of the specified business.

[0011] Further, before the step of inputting the business scene features into a preset knowledge graph to obtain corresponding legal clause features, comprising:

[0012] obtaining multi-source legal data;

[0013] The text features, image features and time sequence features in the multi-source legal data are analyzed by a preset three-channel parallel processing architecture to obtain legal texts;

[0014] Based on a preset intelligent node generation mechanism, subjects in the legal texts and the association relationships between the subjects are extracted to generate a clause-scene mapping model;

[0015] In the clause-scene mapping model, three-dimensional features of quantitative legal attributes, business scenes and regions are constructed to obtain the preset knowledge graph.

[0016] Further, after the step of extracting the subjects in the legal texts and the association relationships between the subjects based on the preset intelligent node generation mechanism to generate the clause-scene mapping model, the method further comprises:

[0017] The legal texts in multiple regions are monitored in real time through a preset API cluster;

[0018] When it is monitored that the legal texts in at least one region have changed, the changed legal texts are obtained;

[0019] Based on the changed legal texts, the corresponding contents in the clause-scene mapping model are incrementally updated.

[0020] Further, the step of inputting the business scene features into the preset knowledge graph to obtain corresponding legal clause features comprises:

[0021] The business scene features are mapped to a preset dimensional semantic space to obtain target business features;

[0022] The target business features are input into the preset knowledge graph to obtain corresponding legal clause features.

[0023] Further, the step of inputting the business parameters into the target cross-border transaction simulator to obtain the cross-border scheme of the specified business comprises:

[0024] Different objective functions are obtained;

[0025] Each of the objective functions and the business parameters is input into the target cross-border transaction simulator to obtain the cross-border scheme of the specified business corresponding to each objective function.

[0026] Further, after the step of inputting each of the objective functions and the business parameters into the target cross-border transaction simulator to obtain the cross-border scheme of the specified business corresponding to each objective function, the method further comprises:

[0027] Obtaining dimension values of compliance, cost-effectiveness, and timeliness of each cross-border scheme;

[0028] Forming a three-dimensional evaluation graph according to the dimension values of compliance, cost-effectiveness, and timeliness;

[0029] Sending the cross-border scheme and the three-dimensional evaluation graph to a designated terminal for selection.

[0030] Further, the step of inputting the business parameters into the target cross-border transaction simulator to obtain the cross-border scheme of the specified business further comprises:

[0031] Real-time monitoring of real-time regional restrictions of each region involved in the cross-border scheme;

[0032] When at least one region has the real-time regional restriction, identifying the warning level of the real-time regional restriction; wherein the warning level is a pre-defined level;

[0033] Sending the warning level and the region with the real-time regional restriction to the target cross-border transaction simulator to generate a real-time alternative cross-border scheme.

[0034] A cross-border scheme generation device based on cross-border rules, the device comprising:

[0035] A business parameter acquisition module for acquiring business scenario features and business parameters of a specified business;

[0036] A legal provision feature acquisition module for inputting the business scenario features into a pre-set knowledge graph to obtain corresponding legal provision features;

[0037] A confidence score acquisition module for obtaining a plurality of relevant cases based on the legal provision features, and performing confidence evaluation on the relevant cases to obtain a confidence score of each relevant case; wherein the relevant cases contain timestamps;

[0038] A standardized case acquisition module for performing pre-set standardization processing on each relevant case based on the confidence score of the relevant case to obtain a standardized case corresponding to each relevant case;

[0039] A target cross-border transaction simulator acquisition module for training a pre-set transaction simulator based on each standardized case to obtain a target cross-border transaction simulator;

[0040] A cross-border scheme acquisition module for inputting the business parameters into the target cross-border transaction simulator to obtain a cross-border scheme of the specified business.

[0041] An electronic device comprises a memory and a processor, the memory storing a computer program, the computer program being executed by the processor to cause the processor to perform the following steps:

[0042] Obtaining a business scenario feature and a business parameter of a specified business;

[0043] Inputting the business scenario feature into a preset knowledge graph to obtain a corresponding legal clause feature;

[0044] Obtaining a plurality of related cases based on the legal clause feature, and performing confidence evaluation on the related cases to obtain a confidence score of each related case; wherein the related cases contain a timestamp;

[0045] Based on the confidence score of the related cases, performing a preset standardization processing on each related case to obtain a corresponding standardized case of each related case;

[0046] Training a preset transaction simulator based on each standardized case to obtain a target cross-border transaction simulator;

[0047] Inputting the business parameter into the target cross-border transaction simulator to obtain a cross-border solution of the specified business.

[0048] A computer-readable storage medium stores a computer program, the computer program being executed by a processor to cause the processor to perform the following steps:

[0049] Obtaining a business scenario feature and a business parameter of a specified business;

[0050] Inputting the business scenario feature into a preset knowledge graph to obtain a corresponding legal clause feature;

[0051] Obtaining a plurality of related cases based on the legal clause feature, and performing confidence evaluation on the related cases to obtain a confidence score of each related case; wherein the related cases contain a timestamp;

[0052] Based on the confidence score of the related cases, performing a preset standardization processing on each related case to obtain a corresponding standardized case of each related case;

[0053] Training a preset transaction simulator based on each standardized case to obtain a target cross-border transaction simulator;

[0054] Inputting the business parameter into the target cross-border transaction simulator to obtain a cross-border solution of the specified business.

[0055] The beneficial effects of the present application: by acquiring the business scene features and business parameters in real time and inputting them into the intelligent knowledge graph, the instant extraction and update of legal clause features are realized, ensuring that the scheme generation always conforms to the latest laws and regulations, the added confidence assessment and standardized processing mechanism efficiently filters and standardizes relevant cases, combined with the target cross-border transaction simulator, enterprises can perform real-time simulation and adjustment when formulating cross-border schemes, greatly improving the flexibility of decision-making. When legal clauses conflict, dynamic coordination can be achieved through intelligent logic analysis, avoiding compliance risks caused by rule conflicts, effectively solving the limitations of traditional rule engines in the face of rapidly changing regional trade agreements and multi-modal regulatory requirements. It has significant technical advantages in improving cross-border trade compliance, flexibility and efficiency, providing more competitive trade strategies and operational management capabilities for enterprises. BRIEF DESCRIPTION OF DRAWINGS

[0056] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0057] Among them:

[0058] Figure 1 An application environment diagram of the cross-border scheme generation method based on cross-border rules in one embodiment;

[0059] Figure 2 A flowchart of the cross-border scheme generation method based on cross-border rules in one embodiment;

[0060] Figure 3 A structural block diagram of the cross-border scheme generation device based on cross-border rules in one embodiment;

[0061] Figure 4 A structural block diagram of an electronic device in one embodiment. DETAILED DESCRIPTION

[0062] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0063] Figure 1 An application environment diagram of the cross-border scheme generation based on cross-border rules in one embodiment. Referring to Figure 1The cross-border scheme generation method based on cross-border rules is applied to a cross-border scheme generation system based on cross-border rules. The cross-border scheme generation system based on cross-border rules includes a terminal 110 and a server 120. The terminal 110 and the server 120 are connected through a network. The terminal 110 can be a desktop terminal or a mobile terminal. The mobile terminal can be at least one of a mobile phone, a tablet computer, a notebook computer, etc. The server 120 can be implemented by an independent server or a server cluster composed of multiple servers. The terminal 110 is used to obtain the business scene features and business parameters of a specified business. The server 120 is used to generate a cross-border scheme for the specified business.

[0064] As shown in Figure 2 In one embodiment, a cross-border scheme generation method based on cross-border rules is provided. The method can be applied to a terminal or a server. In this embodiment, the method is exemplified by being applied to a terminal. The cross-border scheme generation method based on cross-border rules specifically includes the following steps:

[0065] S1: Obtain the business scene features and business parameters of a specified business;

[0066] S2: Input the business scene features into a preset knowledge graph to obtain corresponding legal clause features;

[0067] S3: Obtain a plurality of related cases based on the legal clause features, and perform confidence evaluation on the related cases to obtain a confidence score of each related case. The related cases contain timestamps.

[0068] S4: Based on the confidence score of the related cases, perform preset standardization processing on each related case to obtain a standardized case corresponding to each related case;

[0069] S5: Train a preset transaction simulator based on each standardized case to obtain a target cross-border transaction simulator;

[0070] S6: Input the business parameters into the target cross-border transaction simulator to obtain a cross-border scheme for the specified business.

[0071] As described in step S1 above, the business scenario features and business parameters of the designated business are obtained. The business scenario features can specifically include the purpose, scope, transaction object, market environment, and other aspects of information of the business. In one specific embodiment, the business scenario features include transaction amount, commodity HS code, and payment method, while the business parameters involve specific numerical information, such as transaction amount, transaction frequency, cargo type, payment method, etc. These information can be obtained through various ways such as user input, system detection, or third-party data sources. Accurate collection of these features and parameters is the basis of the entire scheme generation process, as they will directly affect the subsequent matching of legal clauses and retrieval of relevant cases. If some key features cannot be obtained or are incorrect, it will lead to inaccuracy of the scheme, and thus affect the compliance and success rate of the business. Therefore, this step usually needs to design a reasonable interface to facilitate user input and modification of data, and ensure the integrity and accuracy of the data.

[0072] As described in step S2 above, the business scenario features are input into a preset knowledge graph to obtain corresponding legal clause features. The obtained business scenario features are input into a preset knowledge graph, the purpose of which is to extract relevant legal clause features therefrom. The knowledge graph is a structured information storage method that connects different information through relationships, making the information between them have relevance. In the processing of cross-border business, the legal clause features can include the agreement provisions involved, and the most relevant legal clauses need to be matched according to the business scenario. This process usually involves complex natural language processing (NLP) technology and semantic analysis to ensure that the extracted legal clauses truly reflect the needs of the business scenario, while considering the applicability and effectiveness of the legal clauses.

[0073] As described in step S3 above, based on the legal clause features, a plurality of relevant cases are obtained, and the relevant cases are subjected to confidence evaluation to obtain a confidence score of each of the relevant cases. Using the extracted legal clause features, relevant legal cases are queried and obtained, which can come from legal databases, court records, or various legal documents. The obtained cases will involve similar legal background, judgment result, and legal explanation to the current business scenario. Then, the cases are subjected to confidence evaluation so as to assign a confidence score to each case, reflecting the relevance of the case to the current business scenario. This evaluation process considers multiple factors, such as the time stamp of the case, the status of the case, and the application frequency of the case, etc. The high or low of the confidence score will directly affect the result of the subsequent standardization processing, and therefore, detailed analysis and machine learning of the data are needed to ensure the fairness and objectivity of the evaluation process.

[0074] As described in step S4, based on the confidence score of the relevant case, the relevant case is standardized according to a preset standardization process to obtain a standardized case corresponding to the relevant case. Each case is standardized according to its confidence score. The significance of standardization is to convert cases of different types and formats into a unified standard format, making subsequent analysis and comparison easier. The content of standardization can include the structure of the case, the extraction of key elements (such as legal provisions, case summaries, judgment results, etc.), and the consistency of the format. This step requires clear standardization rules to ensure that all cases can be compared on the same basis. At the same time, machine learning algorithms can be used to continuously optimize the standardization process, improving efficiency and accuracy. Through this process, standardized cases will provide high-quality input data for the training objectives of subsequent steps, significantly improving the scientificity and rationality of the generated solutions.

[0075] As described in step S5, based on each standardized case, a preset transaction simulator is trained to obtain a target cross-border transaction simulator. Using the standardized cases, the system will begin training the preset transaction simulator. The core function of the transaction simulator is to simulate and predict the performance of different cross-border transactions, including legal compliance, financial impact, and risk assessment. Through machine learning technology, the training simulator will extract patterns and features from the input cases, learning the most superior transaction strategies in different commercial environments. To ensure the effectiveness of the simulator, a large number of case cross-validation and parameter adjustment are included in the training process to continuously improve its accuracy and flexibility. Ultimately, the target cross-border transaction simulator will form a relatively complete and practical analysis platform, providing accurate transaction recommendations, compliance solutions, and potential risk warnings for users' cross-border business in an efficient and intelligent manner. It should be noted that the preset transaction simulator can be obtained by training a pre-constructed first neural network model based on a preset sample set. Each sample data in the preset sample set includes business parameters of a standardized case and an actual cross-border solution corresponding to the business parameters of the standardized case. When training the pre-constructed first neural network model, the business parameters of the standardized case in each sample data are used as the input of the first neural network model, and the cross-border solution corresponding to the business parameters of the standardized case in each sample data is used as the output of the first neural network model. Through training, the first neural network model can learn the corresponding relationship between all possible business parameters of standardized cases and cross-border solutions. The trained first neural network model is used as the preset transaction simulator. During the training process, the mean square error can be used as the loss function, the stochastic gradient descent can be used as the optimizer for training the first neural network model, the learning rate parameter can be used to control the weight update pace, and finally the root mean square error can be used to evaluate the model to achieve a preset evaluation score.

[0076] The business parameters are input into the target cross-border transaction simulator as described in step S6 to obtain the cross-border solution for the specified business. The user-provided business parameters are input into the trained target cross-border transaction simulator, which means that the user can obtain a cross-border solution generated for their specific situation by simply inputting specific business data such as transaction amount, payment method, etc. The solution will take into account the extracted legal clause features, relevant cases, standardized bases, and reference results of the simulator to ensure compliance, operability, and market adaptability of the solution. The output of this solution can be a detailed compliance report, recommended operation steps, risk assessment, etc., and the user can make corresponding decisions and actions based on these outputs.

[0077] In one embodiment, before the step S2 of inputting the business scenario features into the preset knowledge graph to obtain corresponding legal clause features, the method further comprises:

[0078] S101: Obtain multi-source legal data;

[0079] S102: Analyze the text features, image features, and time sequence features in the multi-source legal data through a preset three-channel parallel processing architecture to analyze and obtain legal texts;

[0080] S103: Extract the subjects and the association relationships between the subjects in the legal texts based on a preset intelligent node generation mechanism to generate a clause-scenario mapping model;

[0081] S104: In the clause-scenario mapping model, construct three-dimensional features of quantitative legal attributes, business scenarios, and regions to obtain the preset knowledge graph.

[0082] As described in step S101, multi-source legal data is obtained. Legal data from different sources is collected, including legal texts, regulations, case judgments, case explanations, and various text types. Multi-source data collection is achieved through methods such as web crawlers, legal database APIs, and information disclosure platforms. To ensure data diversity and reliability, not only higher-level legal documents can be obtained, but also data sources can be enriched from local regulations, industry standards, and legal literature. When obtaining data, attention should be paid to the update frequency and completeness of the data to ensure the timeliness and accuracy of subsequent analysis. By integrating these diverse data sources, a comprehensive and in-depth legal knowledge graph can be constructed in subsequent steps, ensuring wide coverage and strong applicability.

[0083] As described in step S102, the text features, image features, and time sequence features in the multi-source legal data are analyzed by a preset three-channel parallel processing architecture to obtain legal text. In this step, the collected legal data is analyzed using a three-channel parallel processing architecture, which targets text features, image features, and time sequence features. In the text channel, the system uses natural language processing (NLP) techniques to perform word segmentation, part-of-speech tagging, and syntax analysis on legal text to identify legal terminology, key provisions, and their contextual relationships. In the image channel, if there are graphics, tables, or other visual information in the legal data, computer vision (CV) techniques are used to process and interpret the information to extract valuable visual information. In the time sequence channel, the timeliness of legal provisions and the timeline information of provision validity and invalidity are analyzed. Through parallel processing, different features can be analyzed simultaneously, improving the efficiency and effectiveness of data processing, and ultimately forming clear and structured legal text to provide a basis for subsequent extraction of subjects and association relationships. Specifically, in a specific embodiment, the text channel can use a pre-trained legal semantic model to analyze the structure of provisions and identify the validity conditions of provisions; the image channel can apply enhanced OCR technology to process scanned files, especially for fuzzy recognition of various regional seals; and the time sequence channel can track the revision track of provisions by constructing a time axis sequence model.

[0084] As described in step S103, the subjects in the legal text and the association relationships between the subjects are extracted based on a preset intelligent node generation mechanism to generate a provision-scene mapping model. Through the preset intelligent node generation mechanism, significant subjects and association relationships between the subjects are extracted from the processed legal text. In legal text, "subjects" generally refer to parties to legal provisions, such as individuals, companies, and institutions, while association relationships refer to legal relationships and constraints between subject phenomena, such as tax subject, tax item category, tax basis, tax rate standard, and tax administration process. In the extraction process, machine learning models can be used to identify and classify various entities in legal text and construct a network between subjects. Through these extracted subjects and relationships, a "provision-scene mapping model" can be generated, in which legal provisions are combined with specific business scenarios to demonstrate the applicability and influence of legal provisions in different scenarios. This model will provide a scientific data structure for subsequent processing and application of legal provision features and improve the matching accuracy of legal provisions and actual business scenarios. The "intelligent node generation mechanism" can be implemented through knowledge graph extraction algorithms and entity recognition models.

[0085] As described in step S104 above, in the clause-scene mapping model, the three-dimensional characteristics of quantified legal attributes, business scenes, and regions are constructed to obtain the preset knowledge graph. After constructing the clause-scene mapping model, the legal attributes, business scenes, and regional information are quantified to form a three-dimensional feature structure. This process not only requires quantitative analysis of the content of legal clauses, such as the nature, type, and scope of application of legal clauses; but also requires detailed description of business scenes, such as the nature of transactions and the industries involved; and must also consider the differences in regional law, such as the applicability of legal clauses in different regions. Through the integration of these quantified features, a comprehensive and complex knowledge graph can be constructed, connecting the features of the above three dimensions through nodes and edges to form a legal knowledge graph with rich interactive relationships. Such a knowledge graph can not only be used for subsequent legal clause extraction and application, but also can provide intelligent recommendations, legal search, and other functions for users, enhancing the practicality and flexibility of the system.

[0086] In one embodiment, after the step S103 of generating the clause-scene mapping model based on the preset intelligent node generation mechanism to extract the subjects in the legal text and the association relationship between the subjects, the method further comprises:

[0087] S1131: Real-time monitoring of legal texts in multiple regions through a preset API cluster;

[0088] S1132: When it is monitored that there is at least one region where the legal text has changed, obtaining the changed legal text;

[0089] S1133: Incrementally updating the corresponding content in the clause-scene mapping model based on the changed legal text.

[0090] As described in step S1131 above, the legal texts in multiple regions are monitored in real time through a preset API cluster. In this stage, the legal texts in multiple regions are monitored in real time through a preset API cluster. These regional legal texts can be legal texts in various regions, and the monitored content includes legal events such as new legislation, revision, and abolition. By calling the APIs of each region, the latest version of the law and related information can be obtained in a timely manner. This monitoring work usually needs to be combined with data flow technology to obtain fast and efficient update response capability, at the same time, the use of API cluster also makes the monitoring range expand, users can select to monitor specific legal fields or issues according to their own needs, in this stage, it is particularly important to set reasonable monitoring thresholds and conditions to prevent information overload, while ensuring that key legal information can be captured and processed in a timely manner. This real-time monitoring provides an important data basis for subsequent changes in legal clauses.

[0091] As described in step S1132 above, when it is monitored that there is a change in the legal text of at least one region, the changed legal text is obtained. When it is monitored that the legal text of one or more regions is changed, the system will immediately take action to obtain these changed legal texts. This means that once the update of the legal document is identified, the system will activate the change acquisition mechanism to access the latest legal text through the aforementioned API cluster. The changed legal text may involve information such as changes in legal provisions, newly added legal provisions, or amendments to regulations, which are crucial for maintaining the accuracy of the clause-scenario mapping model. When obtaining these updated texts, the system will keep a version of the original text for subsequent comparison and analysis. Through this mechanism, the timeliness and accuracy of legal information can be effectively ensured, and support for dynamic management of legal clauses is provided. Ultimately, this process not only enhances the ability of the law to adapt to changes, but also enhances the flexibility and practicality of the system.

[0092] As described in step S1133 above, the corresponding content in the clause-scenario mapping model is incrementally updated based on the changed legal text. Incremental updating of the clause-scenario mapping model based on these changes involves analyzing the content of the affected parts of the model to determine which elements need to be updated, including legal clauses, subject relationships, and their relevance. Using difference analysis techniques, the system can automatically identify specific modifications in the legal text and adjust the relevant nodes in the clause-scenario mapping model, including adding new clauses, modifying the relevance of existing clauses, or deleting obsolete clauses. At the same time, in order to maintain the consistency of the model, incremental updating also ensures that new legal information is seamlessly integrated into the existing knowledge structure without interfering with the established mapping relationships. Through timely incremental updating, the system can maintain the latest state of the clause-scenario mapping model, thereby ensuring the effectiveness and reliability of the entire knowledge graph in business applications, which is particularly important for decision-making and analysis using legal clauses and scenarios.

[0093] In one embodiment, the step S2 of inputting the business scenario features into the preset knowledge graph to obtain corresponding legal clause features comprises:

[0094] S201: Map the business scenario features to a preset dimensional semantic space to obtain target business features;

[0095] S202: Input the target business features into the preset knowledge graph to obtain corresponding legal clause features.

[0096] As described in step S201 above, the business scenario features are mapped to a pre-set dimensional semantic space to obtain target business features. First, the obtained business scenario features need to be semantically mapped. This process involves defining a pre-set dimensional semantic space, such as projection into a 768-dimensional semantic space. After passing through the input layer and multiple Transformer layers, a 768-dimensional vector is generated, which represents the semantic features of the input word or entire sentence. By mapping the business scenario features to this dimensional space, a more structured and detailed target business feature can be generated, which not only better describes the complexity of the business scenario, but also lays the foundation for subsequent matching with legal provisions.

[0097] This mapping process usually relies on multi-level feature extraction and conversion techniques, such as Word Embedding, Topic Modeling, and other natural language processing techniques, to ensure that the business scenario features are accurately represented and can form an effective corresponding relationship with the legal provision features in the knowledge graph. For different business scenarios, the mapping strategy and dimensions may vary, so the flexibility and adaptability of this process are particularly critical. By generating target business features, the system not only improves the analysis capability of the business scenario, but also provides a clearer context for subsequent extraction of legal provision features.

[0098] As described in step S202 above, the target business features are input into the pre-set knowledge graph to obtain corresponding legal provision features. The target business features are input into the pre-set knowledge graph to obtain corresponding legal provision features. The knowledge graph contains a large amount of legal information, such as legal provisions, cases, regulations, case explanations, etc., which need to be quickly found through the relationships and structures defined in the graph to match the input target business features. The matching process usually relies on image similarity calculation, graph reasoning, and semantic search techniques. For example, according to the keywords extracted from the business features, relevant legal provisions can be found in the knowledge graph, and according to the context semantics, applicability, and regional differences of the provisions, the most relevant legal provision features are selected. At this time, it is not just a simple query, but also needs to consider whether the corresponding legal provisions have timeliness, scope of application, and compliance, and through such a way, the legal provision features obtained finally can accurately reflect the legal norms that the business scenario needs to follow.

[0099] In one embodiment, the step S6 of inputting the business parameters into the target cross-border transaction simulator to obtain the cross-border solution for the specified business includes:

[0100] S601: Obtain different objective functions;

[0101] S602: input each of the target functions and the business parameters into the target cross-border transaction simulator respectively to obtain the cross-border scheme of the designated business corresponding to each target function.

[0102] As described in step S601 above, different target functions are obtained. The system first needs to define and obtain multiple different target functions. These target functions are key indicators used in the target cross-border transaction simulator to evaluate whether the transaction scheme achieves the expected effect. The target functions can be concretized into multiple levels and aspects, such as maximizing revenue, minimizing cost, risk avoidance, compliance evaluation, market share improvement, etc. Each target function may focus on different business objectives or strategies, thereby affecting the decision-making consideration of cross-border transactions. Obtaining target functions not only needs to consider the company's business objectives, but also needs to be designed according to specific business scenarios and market conditions. In actual operation, this step involves market analysis, definition and calculation of financial indicators, determination of compliance standards, etc. For example, in a high-risk market environment, the enterprise may pay more attention to risk management and compliance, while in a highly competitive market, the balance between revenue and cost will become more important. By sorting out different target functions, the simulator can consider a wider perspective and strategy when processing business parameters, thereby providing diversified and comprehensive support for subsequent transaction scheme generation.

[0103] As described in step S602 above, each of the target functions and the business parameters is input into the target cross-border transaction simulator to obtain the cross-border scheme of the designated business corresponding to each target function. For each target function, it is input into the target cross-border transaction simulator together with specific business parameters. This process is the core of realizing dynamic and individualized transaction scheme generation. First, business parameters represent specific transaction information such as amount, product type, payment method, etc., while target functions define the standards required to evaluate a successful transaction. When these information are input into the simulator at the same time, the system will perform calculations and simulations based on built-in algorithms and logic models. Specifically, the target cross-border transaction simulator will use the set strategy optimization technology to generate the corresponding cross-border transaction scheme according to the input target function. For example, if the target function is to maximize profit, the simulator will consider market demand, pricing strategy, cost structure, etc. to propose the best transaction decision and execution path; while if the target is to reduce risk, the simulator will consider compliance measures, market fluctuations, currency risk, etc. to provide more prudent schemes. Finally, the simulator will output multiple cross-border transaction schemes, each of which will be optimized according to different target functions for the enterprise to compare and select. Through this process, the enterprise can obtain adaptive and diversified cross-border transaction schemes, enabling it to effectively adjust strategies and select schemes in a complex and changing market environment, achieving higher business flexibility and response capability.

[0104] In one embodiment, after the step S602 of inputting each of the target functions and the business parameters into the target cross-border transaction simulator to obtain the cross-border scheme of the designated business corresponding to each target function, the method further comprises:

[0105] S6121: obtaining dimension values of compliance, cost-effectiveness, and timeliness of each cross-border scheme;

[0106] S6122: forming a three-dimensional evaluation graph according to the dimension values of compliance, cost-effectiveness, and timeliness;

[0107] S6123: sending the cross-border scheme and the three-dimensional evaluation graph to a designated terminal for selection.

[0108] As described in the step S6121, the dimension values of compliance, cost-effectiveness, and timeliness of each cross-border scheme are obtained. Each cross-border scheme needs to be evaluated in depth, and the values of the three important dimensions of compliance, cost-effectiveness, and timeliness are extracted. These are the key factors for evaluating the success of a cross-border transaction scheme. The compliance dimension mainly considers the degree of compliance of the scheme in terms of laws, regulations, and policies, ensuring that the transaction does not violate the laws and regulations of the target market. This evaluation usually needs to be compared with the preset regulation knowledge base to check whether all relevant provisions are implemented. The cost-effectiveness dimension mainly focuses on the relationship between costs and benefits in the transaction process, evaluating whether the scheme is economically reasonable. Multiple factors such as transaction costs, management fees, and potential benefits are considered, and quantitative analysis is performed through financial models. The timeliness dimension focuses on the timeliness of the transaction in the execution process, including delivery time, response time, and speed of market opportunity grasping, ensuring that the transaction scheme can be completed within a reasonable time frame. By calculating the values of the above three dimensions for each scheme, the system can form a transparent and comprehensive quantitative evaluation of the strategy, providing a solid data foundation for subsequent graphical display and decision-making.

[0109] As described in the step S6122, a three-dimensional evaluation graph is formed according to the dimension values of compliance, cost-effectiveness, and timeliness. Using the three dimension values obtained, a three-dimensional evaluation graph (3D Evaluation Graph) is constructed. This three-dimensional evaluation graph not only makes the data more intuitive, but also clearly shows the relative performance of each scheme in the three dimensions of compliance, cost-effectiveness, and timeliness when comparing multiple cross-border schemes.

[0110] In constructing the three-dimensional graph, compliance, cost-effectiveness and timeliness are taken as the three dimensions of the coordinate axes respectively, and the scheme of each participant will be represented as a point or vector in space. This three-dimensional view can clearly show the advantages and disadvantages between different schemes, helping users intuitively identify which scheme performs better or worse in each dimension. The use of this view not only enhances the readability of information, but also enables the final decision to be based on more scientific and empirical data analysis. Color coding or markers can also be introduced in the graph to further enhance the readability and recognition of information, making users more autonomous and accurate in selecting schemes.

[0111] As described in step S6123 above, the cross-border scheme and the three-dimensional evaluation graph are sent to the designated terminal for selection. All generated cross-border schemes and corresponding three-dimensional evaluation graphs are sent to the user's designated terminal, realizing effective information interaction between the user and the system. The terminal sent may be a mobile device, a computer, or a dedicated business management platform, etc.

[0112] After receiving the cross-border scheme and the evaluation graph, the user can select and compare the schemes through the graphical interface. The intuitive graphical display enables the user to quickly grasp the characteristics of each scheme and its performance differences in multiple dimensions such as compliance, cost-effectiveness and timeliness, and then make a decision that best meets the business strategy and market demand. At the same time, the system can also design interactive functions to enable users to further analyze specific details, such as viewing detailed compliance reports, fee details or timeliness analysis of a scheme, etc. Such flexibility not only improves the user experience, but also greatly enhances the scientificity and effectiveness of the decision-making process, ensuring that the final selected scheme can achieve the best results in actual operation.

[0113] In one embodiment, after the step S6 of inputting the business parameters into the target cross-border transaction simulator to obtain the cross-border scheme of the specified business, the method further comprises:

[0114] S701: Real-time monitoring of real-time regional restrictions of each region involved in the cross-border scheme;

[0115] S702: When at least one region has the real-time regional restriction, identifying the warning level of the real-time regional restriction; wherein the warning level is a pre-defined level;

[0116] S703: Sending the warning level and the region with the real-time regional restriction to the target cross-border transaction simulator to generate a real-time alternative cross-border scheme.

[0117] As described in step S701, real-time regional restrictions of each region involved in the cross-border scheme are monitored in real time. The monitoring of real-time regional restrictions of multiple regions involved in the cross-border scheme is started. This monitoring process is usually achieved through multi-threaded information flow analysis, and the monitoring sources can include websites, announcements, news reports, social media, etc. Real-time regional restrictions can include various information such as trade policies, changes in tax rates, entry and exit controls, market access policies, investment restrictions, etc. By setting up an early warning mechanism, the system can quickly capture relevant information when new policy information is released. This real-time monitoring mechanism can ensure that the system always has the latest policy changes in each region in order to assess the impact of these changes on the cross-border scheme. In addition, the focus of monitoring is not only on the content of the policy itself, but also on the implementation background of the policy and its potential impact. For example, a country adjusts the tax rate due to economic factors, or imposes restrictions on certain goods for some reason.

[0118] As described in step S702, when at least one region has the real-time regional restriction, an early warning level of the real-time regional restriction is identified; wherein the early warning level is a pre-defined level. After monitoring that at least one region has a new real-time regional restriction, the policy is immediately evaluated for the early warning level. The early warning level refers to the risk level that the system assigns to the policy change based on its nature and potential impact. This level will be classified according to pre-set standards, such as high, medium, and low risk levels. Influencing factors can include the magnitude of policy changes on the market, the degree of direct impact on business, the urgency of implementation, historical policy practices, etc. For example, if a region suddenly implements strict import restriction policies, which has a greater impact on cross-border transactions, the system will rate it as high risk; while if the policy adjustment is only a slight adjustment of a tax rate, the risk level may be low. The establishment of this early warning mechanism aims to help users identify potential risk points in a timely manner, thereby providing a basis for subsequent response measures. Identifying the early warning level of real-time regional restrictions can provide a quick and objective reference for decision-makers, helping them to respond flexibly in a rapidly changing environment.

[0119] The early warning level and the region with the real-time regional restriction are sent to the target cross-border transaction simulator to generate a real-time alternative cross-border scheme, as described in step S703. The identified early warning level and affected region information are sent to the target cross-border transaction simulator. Through this process, the features of interactivity and dynamic adjustment are integrated into the cross-border transaction scheme generation system. Once the simulator receives these key information, it will quickly assess the impact using corresponding decision algorithms and models and generate a real-time alternative cross-border scheme based on the latest regional restrictions. This alternative scheme will not be a simple modification. The simulator will quickly analyze the specific impact of policy changes on cross-border transactions and consider how to optimize the transaction path under compliance, such as selecting alternative markets, adjusting the supply chain, optimizing costs, etc. Combined with the objective function, these new schemes should be able to continuously maximize profits while minimizing risks. By linking early warning information with the simulator, a quick response can be made to avoid losses caused by information lag and ensure that cross-border transactions always comply with the most real-time policy requirements. Thus, competitiveness can be maintained in a dynamic market environment and more efficient operation and management can be achieved.

[0120] With reference to Figure 3 The application also provides a cross-border scheme generation device based on cross-border rules, which comprises:

[0121] A business parameter acquisition module 902 is configured to acquire business scene features and business parameters of a specified business.

[0122] A legal clause feature acquisition module 904 is configured to input the business scene features into a preset knowledge graph to acquire corresponding legal clause features.

[0123] A confidence score acquisition module 906 is configured to acquire a plurality of relevant cases based on the legal clause features and perform confidence evaluation on the relevant cases to obtain a confidence score of each relevant case. The relevant cases contain time stamps.

[0124] A standardized case acquisition module 908 is configured to perform preset standardization processing on each relevant case based on the confidence score of the relevant case to obtain a standardized case corresponding to each relevant case.

[0125] A target cross-border transaction simulator acquisition module 910 is configured to train a preset transaction simulator based on each standardized case to obtain a target cross-border transaction simulator.

[0126] A cross-border scheme acquisition module 912 is configured to input the business parameters into the target cross-border transaction simulator to obtain a cross-border scheme of the specified business.

[0127] In an embodiment, the cross-border scheme generation device based on cross-border rules further comprises:

[0128] The multi-source legal data acquisition module is configured to acquire multi-source legal data.

[0129] The legal text acquisition module is configured to analyze text features, image features, and timing features in the multi-source legal data through a preset three-channel parallel processing architecture to obtain legal text.

[0130] The clause-scene mapping model generation module is configured to extract subjects in the legal text and association relationships between the subjects based on a preset intelligent node generation mechanism to generate a clause-scene mapping model.

[0131] The preset knowledge graph acquisition module is configured to construct three-dimensional features of quantified legal attributes, business scenes, and regions in the clause-scene mapping model to obtain the preset knowledge graph.

[0132] In an embodiment, the cross-border scheme generation device based on cross-border rules further comprises:

[0133] The legal text monitoring module is configured to monitor legal texts in multiple regions in real time through a preset API cluster.

[0134] The changed legal text acquisition module is configured to acquire changed legal texts when it is monitored that legal texts in at least one region have changed.

[0135] The incremental update module is configured to perform incremental update on corresponding content in the clause-scene mapping model based on the changed legal texts.

[0136] In an embodiment, the legal clause feature acquisition module 904 comprises:

[0137] The target business feature acquisition submodule is configured to map the business scene features to a preset dimensional semantic space to obtain target business features.

[0138] The legal clause feature acquisition submodule is configured to input the target business features into a preset knowledge graph to obtain corresponding legal clause features.

[0139] In an embodiment, the cross-border scheme acquisition module 912 comprises:

[0140] The target function acquisition submodule is configured to acquire different target functions.

[0141] The cross-border scheme acquisition submodule is configured to input each target function and the business parameter into the target cross-border transaction simulator to obtain a cross-border scheme of the specified business corresponding to each target function.

[0142] In one embodiment, the cross-border solution acquisition module 912 further includes:

[0143] The dimensional value acquisition submodule is used to obtain the dimensional values ​​of compliance, cost-effectiveness, and timeliness for each cross-border solution.

[0144] The 3D evaluation chart generation submodule is used to generate a 3D evaluation chart based on the dimensional values ​​of compliance, cost-effectiveness, and timeliness.

[0145] The sending submodule is used to send the cross-border solution and the three-dimensional evaluation map to a designated terminal for selection.

[0146] In one embodiment, the cross-border scheme generation apparatus based on cross-border rules further includes:

[0147] The real-time area restriction monitoring module is used to monitor the real-time area restrictions of each area involved in the cross-border scheme.

[0148] The warning level identification module is used to identify the warning level of the real-time area restriction when at least one area has the real-time area restriction; wherein the warning level is a pre-defined level;

[0149] The real-time alternative cross-border solution generation module is used to send the warning level and the region with the real-time regional restriction to the target cross-border transaction simulator to generate a real-time alternative cross-border solution.

[0150] Figure 4 An internal structural diagram of an electronic device in one embodiment is shown. This electronic device can specifically be a terminal or a server, and more specifically, a computer device. Figure 4 As shown, the electronic device includes a processor, a memory, and a network interface connected via a system bus. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and may also store a computer program. When executed by the processor, this computer program enables the processor to implement a cross-border scheme generation method based on cross-border rules. The internal memory may also store a computer program, which, when executed by the processor, enables the processor to execute a cross-border scheme generation method based on cross-border rules. Those skilled in the art will understand that... Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0151] In one embodiment, an electronic device is provided, comprising a memory and a processor, the memory storing a computer program, the computer program being executed by the processor to cause the processor to perform the following steps:

[0152] obtaining a business scenario feature and a business parameter of a specified business;

[0153] inputting the business scenario feature into a preset knowledge graph to obtain a corresponding legal clause feature;

[0154] obtaining a plurality of related cases based on the legal clause feature, and performing confidence evaluation on the related cases to obtain a confidence score of each related case; wherein the related cases contain a timestamp;

[0155] based on the confidence score of the related cases, performing a preset standardization processing on each related case to obtain a corresponding standardized case of each related case;

[0156] training a preset transaction simulator based on each standardized case to obtain a target cross-border transaction simulator;

[0157] inputting the business parameter into the target cross-border transaction simulator to obtain a cross-border solution of the specified business.

[0158] By obtaining the business scenario feature and the business parameter in real time and inputting them into the intelligent knowledge graph, the instant extraction and update of the legal clause feature are realized, ensuring that the solution generation always complies with the latest laws and regulations. The added confidence evaluation and standardization processing mechanism efficiently filters and standardizes related cases. Combined with the target cross-border transaction simulator, enterprises can perform real-time simulation and adjustment when formulating cross-border solutions, greatly improving the flexibility of decision-making. In the event of conflicts in legal clauses, dynamic coordination can be achieved through intelligent logic analysis, avoiding compliance risks caused by rule conflicts, and effectively solving the limitations of traditional rule engines in the face of rapidly changing regional trade agreements and multi-modal regulatory requirements. In terms of improving cross-border trade compliance, flexibility and efficiency, it has significant technical advantages, providing more competitive trade strategies and operational management capabilities for enterprises.

[0159] In one embodiment, a computer-readable storage medium is provided, storing a computer program, the computer program being executed by a processor to cause the processor to perform the following steps:

[0160] obtaining a business scenario feature and a business parameter of a specified business;

[0161] inputting the business scenario feature into a preset knowledge graph to obtain a corresponding legal clause feature;

[0162] obtain a plurality of relevant cases based on the legal provision features, and perform confidence evaluation on the relevant cases to obtain a confidence score of each of the relevant cases; wherein the relevant cases contain time stamps;

[0163] perform preset standardization processing on each of the relevant cases based on the confidence score of the relevant cases to obtain a standardized case corresponding to each of the relevant cases;

[0164] train a preset transaction simulator based on each of the standardized cases to obtain a target cross-border transaction simulator;

[0165] input the business parameters into the target cross-border transaction simulator to obtain a cross-border solution for the specified business.

[0166] By acquiring business scene features and business parameters in real time and inputting them into the intelligent knowledge graph, instant extraction and updating of legal provision features are realized, ensuring that the solution generation always complies with the latest laws and regulations. The added confidence evaluation and standardization processing mechanism efficiently filters and standardizes relevant cases. Combined with the target cross-border transaction simulator, enterprises can perform real-time simulation and adjustment when formulating cross-border solutions, greatly improving the flexibility of decision-making. In the event of conflicts in legal provisions, dynamic coordination can be achieved through intelligent logic analysis, avoiding compliance risks caused by rule conflicts and effectively addressing the limitations of traditional rule engines in the face of rapidly changing regional trade agreements and multi-modal regulatory requirements. In terms of improving cross-border trade compliance, flexibility and efficiency, the system has significant technical advantages, providing enterprises with more competitive trade strategies and operational management capabilities.

[0167] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The program can be stored in a non-volatile computer readable storage medium, and when the program is executed, the processes of the above-mentioned embodiment methods can be included. Any reference to memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0168] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0169] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent of the present application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A cross-border scheme generation method based on cross-border rules, characterized in that, The method comprises: obtaining business scene characteristics and business parameters of a specified business; obtaining multi-source legal data; analyzing text features, image features and time sequence features in the multi-source legal data through a preset three-channel parallel processing architecture to obtain legal texts; extracting subjects in the legal texts and the association relationship between the subjects based on a preset intelligent node generation mechanism to generate a clause-scene mapping model; constructing three-dimensional features of quantified legal attributes, business scenes and regions in the clause-scene mapping model to obtain a preset knowledge graph; inputting the business scene characteristics into the preset knowledge graph to obtain corresponding legal clause characteristics; obtaining a plurality of related cases based on the legal clause characteristics, and performing confidence evaluation on the related cases to obtain confidence scores of the related cases; wherein the related cases contain timestamps; performing preset standardization processing on the related cases based on the confidence scores of the related cases to obtain standardized cases corresponding to the related cases; training a preset transaction simulator based on the standardized cases to obtain a target cross-border transaction simulator; inputting the business parameters into the target cross-border transaction simulator to obtain a cross-border solution for the specified business; The step of inputting the business parameters into the target cross-border transaction simulator to obtain the cross-border solution for the specified business comprises: obtaining different objective functions; inputting each of the objective functions and the business parameters into the target cross-border transaction simulator to obtain a cross-border solution for the specified business corresponding to each objective function; obtaining dimension values of three dimensions of compliance, cost-effectiveness and timeliness of each cross-border solution; forming a three-dimensional evaluation graph according to the dimension values of the three dimensions of compliance, cost-effectiveness and timeliness; sending the cross-border solution and the three-dimensional evaluation graph to a specified terminal for selection. 2.The cross-border scheme generation method based on cross-border rules according to claim 1, characterized in that, After the step of extracting subjects in the legal texts and the association relationship between the subjects based on a preset intelligent node generation mechanism to generate a clause-scene mapping model, the method further comprises: monitoring legal texts in multiple regions in real time through a preset API cluster; when it is monitored that legal texts in at least one region have changed, obtaining the changed legal texts; based on the changed legal texts, incrementally updating corresponding contents in the clause-scene mapping model. 3.The cross-border scheme generation method based on cross-border rules according to claim 1, characterized in that, The step of inputting the business scene characteristics into the preset knowledge graph to obtain corresponding legal clause characteristics comprises: mapping the business scene characteristics to a preset dimensional semantic space to obtain target business characteristics; inputting the target business characteristics into the preset knowledge graph to obtain corresponding legal clause characteristics. 4.The cross-border scheme generation method based on cross-border rules according to claim 1, characterized in that, After the step of inputting the business parameters into the target cross-border transaction simulator to obtain the cross-border solution for the specified business, the method further comprises: monitoring real-time regional restrictions of each region involved in the cross-border solution in real time; when at least one region has the real-time regional restrictions, identifying the warning level of the real-time regional restrictions; wherein the warning level is a pre-defined level. The pre-warning level and the region with the real-time regional restriction are sent to the target cross-border transaction simulator to generate a real-time alternative cross-border solution.

5. A cross-border scheme generating apparatus based on cross-border rules, characterized by, The device comprises: a business parameter acquisition module configured to acquire business scenario characteristics and business parameters of a specified business; a multi-source legal data acquisition module configured to acquire multi-source legal data; a legal text acquisition module configured to analyze text features, image features, and timing features in the multi-source legal data through a preset three-channel parallel processing architecture to obtain legal texts; a clause-scenario mapping model generation module configured to extract subjects and associated relationships between the subjects in the legal texts based on a preset intelligent node generation mechanism to generate a clause-scenario mapping model; a preset knowledge graph acquisition module configured to construct three-dimensional features of quantified legal attributes, business scenarios, and regions in the clause-scenario mapping model to obtain the preset knowledge graph; a legal clause feature acquisition module configured to input the business scenario characteristics into the preset knowledge graph to obtain corresponding legal clause features; a confidence score acquisition module configured to acquire a plurality of relevant cases based on the legal clause features and perform confidence evaluation on the relevant cases to obtain confidence scores of the relevant cases; wherein the relevant cases contain timestamps; a standardized case acquisition module configured to perform preset standardization processing on the relevant cases based on the confidence scores of the relevant cases to obtain standardized cases corresponding to the relevant cases; a target cross-border transaction simulator acquisition module configured to train a preset transaction simulator based on the standardized cases to obtain a target cross-border transaction simulator; a cross-border solution acquisition module configured to input the business parameters into the target cross-border transaction simulator to obtain a cross-border solution for the specified business; the cross-border solution acquisition module comprises: a target function acquisition submodule configured to acquire different target functions; a cross-border solution acquisition submodule configured to input each of the target functions and the business parameters into the target cross-border transaction simulator to obtain cross-border solutions for the specified business corresponding to each of the target functions; a dimension value acquisition submodule configured to acquire dimension values of compliance, cost-effectiveness, and timeliness of each cross-border solution; a three-dimensional evaluation graph formation submodule configured to form a three-dimensional evaluation graph according to the dimension values of compliance, cost-effectiveness, and timeliness; a sending submodule configured to send the cross-border solutions and the three-dimensional evaluation graph to a specified terminal for selection.

6. A computer-readable storage medium, characterized in that, A computer program is stored, and when executed by a processor, causes the processor to perform the steps of the cross-border solution generation method based on cross-border rules according to any one of claims 1 to 4.

7. An electronic device, comprising: The device comprises a memory and a processor, and the memory stores a computer program which, when executed by the processor, causes the processor to perform the steps of the cross-border solution generation method based on cross-border rules according to any one of claims 1 to 4.

Images