Construction method and system of cross-border comprehensive service platform

Abstract

The invention relates to a cross-border comprehensive service platform construction method and system, and the method comprises the steps: obtaining system architecture data and service management information of a cross-border service platform, carrying out the module division of the system architecture data based on the service management information, and obtaining the corresponding core service module information; acquiring real-time processing data of the cross-border service platform, and performing association processing on the real-time processing data and the core service module information to obtain a corresponding real-time service processing group; acquiring security data of the cross-border service platform, and performing security prediction on the core service module information according to the security data to obtain a corresponding security protection trend; a corresponding exception prediction processing strategy is obtained; and performing comprehensive analysis on the initial platform deployment scheme and the exception prediction processing strategy to obtain a corresponding global service optimization strategy. According to the method, the response capability of the platform to dynamic business requirements can be remarkably enhanced, and efficient processing of cross-border services is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of integrated service platforms, and in particular to a method and system for constructing a cross-border integrated service platform. Background Art

[0002] As an important support system for modern international trade and cross-border business, the cross-border comprehensive service platform plays a key role in promoting the process of global economic integration. With the continuous expansion of the scale of cross-border trade and the increasing complexity of business forms, how to build a safe, efficient and reliable cross-border service platform to achieve all-round business support and risk prevention and control has become one of the key directions of current research. The current cross-border service platform construction methods mainly have the following problems: the existing platform architecture design often pays too much attention to the realization of a single business function, ignoring the correlation and synergy between different service modules; the platform deployment plan is usually formulated based on experience, and fails to fully consider the balance between real-time business needs and security protection, resulting in insufficient overall service efficiency of the platform. These problems seriously restrict the improvement of the security performance and service quality of the cross-border service platform. Summary of the invention

[0003] The main purpose of the present invention is to provide a method and system for constructing a cross-border integrated service platform, which can significantly enhance the platform's responsiveness to dynamic business needs and ensure efficient processing of cross-border services.

[0004] To achieve the above-mentioned purpose, the present invention provides a method for constructing a cross-border integrated service platform, comprising: Acquire system architecture data and service management information of the cross-border service platform, divide the system architecture data into modules based on the service management information, and obtain corresponding core service module information; Acquire the real-time processing data of the cross-border service platform, associate the real-time processing data with the core service module information, and obtain a corresponding real-time service processing group; Acquire the security data of the cross-border service platform, perform security prediction on the core service module information based on the security data, and obtain corresponding security protection trends; Performing service planning analysis on the real-time service processing group and the security protection trend to obtain a corresponding initial platform deployment plan; Performing abnormality identification on the security data to obtain corresponding abnormality information; Input the abnormal information and the security protection trend into a preset multi-objective security model to perform MOSMPC strategy analysis to obtain a corresponding abnormal prediction and processing strategy; The initial platform deployment plan and the abnormal prediction and processing strategy are comprehensively analyzed to obtain a corresponding global service optimization strategy.

[0005] Furthermore, the system architecture data and service management information of the cross-border service platform are obtained, and the system architecture data is divided into modules based on the service management information to obtain corresponding core service module information, including: Collecting the architecture of the cross-border service platform to obtain original architecture data; Performing redundancy optimization on the original architecture data to obtain the system architecture data; Acquire the service management information of the cross-border service platform, and perform functional relationship analysis on the system architecture data to obtain a module functional relationship diagram; According to the module function relationship diagram, the system architecture data is hierarchically divided to obtain hierarchical module information; Performing performance evaluation on the hierarchical module information to obtain corresponding performance evaluation results; According to the performance evaluation result and the module function relationship diagram, the hierarchical module information is prioritized to obtain corresponding module priority sorting information; A core module analysis is performed based on the module priority sorting information and the service management information to obtain the core service module information.

[0006] Furthermore, the real-time processing data of the cross-border service platform is obtained, and the real-time processing data is associated with the core service module information to obtain a corresponding real-time service processing group, including: Collecting real-time data streams from the cross-border service platform to obtain the real-time processing data; Performing redundancy analysis on the real-time processing data according to the core service module information to obtain a redundant service relationship list; The redundant service relationship list is subjected to a dynamic analysis of service call frequency and user demand to obtain optimization module information.

[0007] Performing preliminary correlation matching on the optimization module information and the real-time processing data set to obtain an initial real-time processing group; Calculating the index of the initial real-time processing group to obtain corresponding initial service index data; Optimizing the service configuration of the initial service indicator data according to the redundant service relationship list to obtain a corresponding optimized service configuration; The initial real-time processing group is optimized according to the optimized service configuration to obtain the real-time service processing group.

[0008] Furthermore, the obtaining of the security data of the cross-border service platform and the security prediction of the core service module information based on the security data to obtain the corresponding security protection trend include: Collecting log records of the cross-border service platform to obtain original security log data; Performing event classification on the original security log data to obtain classified security event data; Performing time series analysis on the classified security event data to obtain a time series security event sequence; Perform frequency distribution statistics according to the time-series security event sequence to obtain event frequency distribution data; Extracting features from the event frequency distribution data to obtain a security threat feature set; Perform module matching on the core service module information according to the security threat feature set to obtain a security-related module list; Performing a security assessment on the security-related module list to obtain module security score data; Perform scenario prediction on the event frequency distribution data according to the module safety score data to obtain initial protection scenario data; Based on the security threat feature set, trend analysis is performed on the initial protection scenario data to obtain a security protection trend.

[0009] Furthermore, the service planning analysis is performed on the real-time service processing group and the security protection trend to obtain a corresponding initial platform deployment solution, including: Performing data flow analysis on the real-time service processing group to obtain service load distribution data; Performing resource demand statistics on the service load distribution data to obtain platform resource demand data; Dividing the security protection trend into security domains to obtain security protection area data; Perform topology optimization according to the platform resource demand data and the security protection area data to obtain an initial topology solution; Performing service capacity evaluation on the initial topology structure solution to obtain service capacity evaluation data; Perform load balancing on the initial topology scheme according to the service capacity evaluation data to obtain a load balancing deployment scheme; Performing redundancy calculation on the load balancing deployment scheme to obtain a service redundancy configuration scheme; The service redundancy configuration scheme is optimized for resource scheduling to obtain the initial platform deployment scheme.

[0010] Furthermore, the abnormality identification is performed on the security data to obtain corresponding abnormality information, and the MOSMPC strategy analysis is performed on the security protection trend and the abnormality information through a preset multi-objective security model to obtain a corresponding abnormality prediction and processing strategy, including: Extracting data features from the security data to obtain a corresponding security feature vector group; Performing cluster analysis on the security feature vector group to obtain corresponding feature cluster center points; Performing density calculation on the security feature vector group according to the feature cluster center point to obtain a corresponding vector density distribution map; Performing threshold segmentation on the vector density distribution map to obtain corresponding abnormal candidate regions; An abnormal traversal detection is performed on the security feature vector group according to the abnormal candidate area to obtain the abnormal information.

[0011] Performing time series decomposition processing on the safety protection trend to obtain a corresponding trend feature matrix; Perform multi-dimensional feature mapping on to obtain the corresponding abnormal feature matrix; The abnormal information and the abnormal feature matrix are input into the preset multi-objective safety model, and MOSMPC iterative processing is performed to obtain the abnormal prediction processing strategy.

[0012] Furthermore, the abnormal information and the security protection trend are input into a preset multi-objective security model for MOSMPC strategy analysis to obtain a corresponding abnormal prediction and processing strategy, including: Inputting the abnormal information into the multi-objective security model, performing an overall security situation assessment on the abnormal information through the system security layer of the multi-objective security model, and obtaining system-level security situation information; Performing business process security analysis and key application risk assessment on the abnormal information through the application security layer of the multi-objective security model to obtain application layer security risk information; Inputting the security protection trend into the multi-objective security model, performing data flow trajectory and sensitive data distribution analysis on the security protection trend through the data security layer of the multi-objective security model, and obtaining data security situation information; Through the network security layer of the multi-objective security model, network topology feature analysis and abnormal traffic identification are performed on the security protection trend to obtain network security status information; Performing physical device status assessment and boundary protection analysis on the security protection trend through the physical security layer of the multi-objective security model to obtain physical security boundary information; Through the policy processing layer of the multi-objective security model, function constraints are constructed on the system-level security situation information, the application-layer security risk information, the data security situation information, the network security status information and the physical security boundary information to obtain a corresponding MOSMPC constraint condition set function; Solving sub-problems of the MOSMPC constraint condition set function to obtain a corresponding distributed solution; Dynamically adjusting and solving sub-problems of the distributed solution to obtain a local target solution; Performing global coordinated optimization on the local objective solution and the distributed solution to obtain an initial solution; The constraint parameters of the initial solution are iterated to obtain and output the abnormality prediction processing strategy.

[0013] Furthermore, the initial platform deployment scheme and the abnormal prediction and processing strategy are comprehensively analyzed to obtain a corresponding global service optimization strategy, including: Decomposing the initial platform deployment plan hierarchically to obtain corresponding multi-layer deployment architecture data; Performing hierarchical mapping on the anomaly prediction and processing strategy according to the multi-layer deployment architecture data to obtain a corresponding multi-dimensional risk assessment matrix; Performing weighted fusion on the multi-layer deployment architecture data according to the multi-dimensional risk assessment matrix to obtain a corresponding fusion service architecture; Performing dynamic optimization calculation on the fusion service architecture to obtain corresponding timing optimization parameters; Iteratively optimize the fusion service architecture according to the timing optimization parameters to obtain a corresponding optimized service solution; Verifying the constraints of the optimization service solution to obtain corresponding verification result data; The optimized service solution is modified according to the verification result data to obtain a corresponding global service optimization strategy.

[0014] The present invention also provides a system for constructing a cross-border integrated service platform, which is applied to any of the above-mentioned methods for constructing a cross-border integrated service platform, comprising: A collection module, which is used to obtain system architecture data and service management information of the cross-border service platform, divide the system architecture data into modules based on the service management information, and obtain corresponding core service module information; An analysis module, the analysis module is used to obtain real-time processing data of the cross-border service platform, associate the real-time processing data with the core service module information, and obtain a corresponding real-time service processing group; An association module, the association module is used to obtain the security data of the cross-border service platform, perform security prediction on the core service module information based on the security data, and obtain a corresponding security protection trend; A processing module, the processing module is used to perform service planning analysis on the real-time service processing group and the security protection trend to obtain a corresponding initial platform deployment plan; A control module, the control module is used to identify abnormalities of the security data and obtain corresponding abnormal information; An execution module, wherein the execution module is used to input the abnormal information and the security protection trend into a preset multi-objective security model to perform MOSMPC strategy analysis to obtain a corresponding abnormal prediction and processing strategy; An analysis module is used to comprehensively analyze the initial platform deployment plan and the abnormal prediction and processing strategy to obtain a corresponding global service optimization strategy.

[0015] The present invention provides a method and system for constructing a cross-border integrated service platform, which has the following beneficial effects: By associating real-time processing data with core service modules to form a real-time service processing group, the platform's ability to respond to dynamic business needs is significantly enhanced, ensuring efficient processing of cross-border services. The introduction and predictive analysis of security data enable the platform to have the ability to perceive potential security threats in advance, and provide more accurate risk warnings in combination with security protection trends, providing guarantees for the stable operation of cross-border businesses. By combining abnormal information with multi-objective security models, an abnormal prediction and processing strategy is generated, which can effectively solve the problem of rapid response under dynamic security threats. The comprehensive optimization analysis of the initial platform deployment plan and security strategy enables the platform's resource allocation and service performance to achieve the optimal balance of overall service efficiency while meeting business needs, thereby reducing system operating costs and improving user experience. This global optimization strategy provides important support for building an efficient, secure, and intelligent cross-border service platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flowchart of a method for constructing a cross-border integrated service platform is provided for the present invention; Figure 2 This is a system structure diagram for constructing a cross-border comprehensive service platform provided by the present invention.

[0017] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods.

[0020] Reference Figure 1As shown, the present invention provides a method for constructing a cross-border integrated service platform, comprising: Step S1: Obtain system architecture data and service management information of the cross-border service platform, divide the system architecture data into modules based on the service management information, and obtain corresponding core service module information; Step S2: Acquire the real-time processing data of the cross-border service platform, associate the real-time processing data with the core service module information, and obtain the corresponding real-time service processing group; Step S3: Obtain security data of the cross-border service platform, perform security prediction on the core service module information based on the security data, and obtain corresponding security protection trends; Step S4: Perform service planning analysis on the real-time service processing group and security protection trend to obtain the corresponding initial platform deployment plan; Step S5: Identify abnormalities in the security data and obtain corresponding abnormal information; Step S6: Input the abnormal information and security protection trend into the preset multi-objective security model to perform MOSMPC strategy analysis to obtain the corresponding abnormal prediction and processing strategy; Step S7: Comprehensively analyze the initial platform deployment plan and the exception prediction and processing strategy to obtain the corresponding global service optimization strategy.

[0021] Based on the above steps, the detailed process is as follows: Step S1: Comprehensively collect the system architecture data of the cross-border service platform, including infrastructure information such as hardware infrastructure configuration, network topology, server distribution, and data storage architecture. Also obtain the service management information of the platform, which includes management-level information such as business process definition, service level agreement (SLA), resource scheduling strategy, and load balancing rules. Based on these service management information as the basis for division, the system architecture data is modularized. In specific implementation, the platform can be decoupled according to business functions and divided into different functional modules by adopting the service-oriented architecture (SOA) idea. For example, it can be divided into core functional modules such as user authentication module, order processing module, payment settlement module, logistics tracking module, and customs declaration document module. This module division needs to consider factors such as the dependency relationship between modules, interface definition, and data flow to ensure that each module is relatively independent and can work closely together. Through such module division, the maintainability and scalability of the system are improved, providing a basis for subsequent service optimization.

[0022] Step S2: Collect real-time processing data generated during the operation of the platform. These data include runtime data such as user access logs, transaction processing records, system resource usage, network traffic data, and service response time. Map and associate these real-time data with the core service modules divided in step S1, and analyze the real-time operation status and performance of each module. Through data association analysis, key information such as the call relationship, load distribution, and performance bottlenecks between different service modules are identified. Based on these analysis results, the associated service modules are organized into real-time service processing groups for unified monitoring and optimization. For example, related modules such as order processing, payment processing, and logistics processing can be organized into a transaction processing group to achieve collaborative optimization of these modules. This grouping method helps to better understand and manage the interactive relationship between services, and provides a basis for subsequent performance optimization and resource scheduling.

[0023] Step S3: Based on the acquisition of the real-time service processing group, security-related data is collected, including access control logs, security audit records, threat detection logs, vulnerability scanning reports, security event records and other security-related data. These security data will be combined with the core service module information for analysis, and the security risks that each service module may face will be predicted through machine learning or other predictive analysis methods. Predictive analysis will consider information from multiple dimensions such as historical security events, current security situation, and emerging threat types, and generate security protection trend reports for different service modules. This trend report may include potential attack risk assessments, security vulnerability evolution trends, and effectiveness predictions of protective measures. Through such security predictive analysis, the platform can identify potential security threats in advance and provide a basis for the formulation of subsequent security protection strategies, ensuring that the platform can proactively respond to various security challenges.

[0024] Step S4: Based on the real-time service processing group data and security protection trend information obtained in the previous step, the service planning analysis method is used to formulate the initial platform deployment plan. The analysis process first integrates the performance indicators, resource utilization, service response time and other operating data of the real-time service processing group, and combines the risk points and protection requirements predicted in the security protection trend to comprehensively evaluate the deployment requirements of the platform. On this basis, a specific deployment plan including server configuration, network architecture, storage plan, security facilities and other elements is formulated. For example, for high-frequency access service modules, distributed deployment and load balancing strategies can be adopted; for security-sensitive modules, independent security protection measures and backup mechanisms are configured. At the same time, factors such as business scale expansion, peak load processing, and disaster recovery backup are considered to ensure that the deployment plan has sufficient scalability and reliability. The plan also includes specific resource allocation strategies, service quality assurance measures, monitoring and alarm mechanisms and other supporting content.

[0025] Step S5: Conduct in-depth anomaly identification and analysis on the collected security data, and use multi-dimensional anomaly detection methods to discover potential security threats. The anomaly identification process combines statistical analysis, machine learning, rule matching and other technical means to analyze from multiple dimensions such as network traffic characteristics, access behavior patterns, system call sequences, resource usage, etc. By establishing a normal behavior baseline, identify abnormal events that deviate from the normal range, such as abnormal login attempts, suspicious data access, abnormal resource usage, abnormal network connections, etc. The identified anomaly information contains detailed feature descriptions such as anomaly type, occurrence time, impact range, severity, etc. At the same time, the identified anomalies are associated with analysis to discover possible attack chains and threat paths, form a complete abnormal information portrait, and provide a decision-making basis for subsequent anomaly prediction and processing.

[0026] Step S6: Input the obtained abnormal information and security protection trend data into the Multi-Objective Security Model with Predictive Control (MOSMPC) for analysis. The multi-objective security model comprehensively optimizes the abnormal prediction and processing strategy by setting multiple optimization goals such as security, performance impact, and resource consumption. The model first extracts features and recognizes patterns from the abnormal information, and builds a dynamic threat assessment model based on the prediction results of the security protection trend. Based on the assessment results, a processing plan is generated at multiple levels including preventive measures, real-time response, and recovery strategies. The strategy content covers specific implementation plans such as security rule configuration, resource scheduling strategy, emergency response process, and data protection measures, while considering the synergy and resource constraints between different strategies to ensure the feasibility and effectiveness of the strategy.

[0027] Step S7: Integrate the initial platform deployment plan and the anomaly prediction and handling strategy, and conduct a global service optimization analysis. The analysis process evaluates the matching and synergy between the initial deployment plan and the anomaly handling strategy based on multiple dimensions such as service quality indicators, security protection effects, and resource utilization efficiency. Adjust the potential conflicts and optimization spaces found in the evaluation, such as re-planning resource allocation, optimizing the service deployment structure, and enhancing security protection measures. The global optimization strategy integrates considerations at multiple levels such as business needs, performance requirements, and security assurance to form a unified service quality assurance system. The optimization strategy also includes a dynamic adjustment mechanism that can adaptively adjust service configurations and protection measures based on actual operating conditions and changes in security situations to ensure continuous optimization and safe operation of the platform.

[0028] The present invention provides a method and system for constructing a cross-border integrated service platform, which has the following beneficial effects: By associating real-time processing data with core service modules to form a real-time service processing group, the platform's ability to respond to dynamic business needs is significantly enhanced, ensuring efficient processing of cross-border services. The introduction and predictive analysis of security data enable the platform to have the ability to perceive potential security threats in advance, and provide more accurate risk warnings in combination with security protection trends, providing guarantees for the stable operation of cross-border businesses. By combining abnormal information with multi-objective security models, an abnormal prediction and processing strategy is generated, which can effectively solve the problem of rapid response under dynamic security threats. The comprehensive optimization analysis of the initial platform deployment plan and security strategy enables the platform's resource allocation and service performance to achieve the optimal balance of overall service efficiency while meeting business needs, thereby reducing system operating costs and improving user experience. This global optimization strategy provides important support for building an efficient, secure, and intelligent cross-border service platform.

[0029] In one embodiment, the system architecture data and service management information of the cross-border service platform are obtained, and the system architecture data is divided into modules based on the service management information to obtain corresponding core service module information, including: When obtaining the system architecture data and service management information of the cross-border integrated service platform, the architecture of the cross-border service platform is collected through the system architecture collection unit, and the collected content includes server hardware configuration information, network topology information, database structure information, and application deployment information, etc., to form the original architecture data. The original architecture data is optimized based on the data redundancy optimization rules, and the data redundancy optimization rules include duplicate data deletion, data structure simplification, data correlation analysis, etc., and the system architecture data is obtained through optimization processing.

[0030] When obtaining service management information of the cross-border service platform, various management data generated during the operation of the platform are collected, including user access logs, system performance monitoring data, service call records, etc. Functional relationship analysis is performed on the system architecture data, and the analysis content involves the call relationship between various functional modules, data flow process, business logic association, etc., and a module function relationship diagram is drawn. The relationship diagram clearly shows the hierarchical relationship and data interaction method between modules.

[0031] According to the module function relationship diagram, the system architecture data is hierarchically divided according to the functional hierarchical division rules, including business logic level, data processing level, user interaction level, etc., to generate hierarchical module information. The hierarchical module information is evaluated for performance, and the evaluation indicators include response time, concurrent processing capability, resource occupancy rate, error rate and other parameters. The performance evaluation results are calculated through the performance evaluation model.

[0032] According to the established module priority assessment criteria, combined with the performance evaluation results and the module function relationship diagram, the hierarchical module information is prioritized. The assessment criteria cover factors such as module importance, frequency of use, and degree of performance impact. The sorting process adopts a multi-dimensional scoring method to ultimately generate module priority information. In the core module analysis phase, the module priority information is correlated with the service management information, and based on the core function identification rules, the modules that play a key role in the system operation are identified, and the core service module information is ultimately determined. The core function identification rules comprehensively consider multiple dimensions such as the module's business value, system dependency, and performance impact.

[0033] This embodiment achieves efficient management and utilization of platform architecture data by systematically collecting and optimizing the architecture of the cross-border service platform. The module function relationship analysis based on service management information accurately shows the hierarchical relationship and data interaction mode between modules, providing a reliable basis for subsequent module division. The multi-dimensional scoring method is used to prioritize the modules, effectively identify the key modules of the system, and improve the efficiency of system resource allocation. In the core module analysis stage, the core service module information is determined through association analysis, so that the system performance is significantly optimized and the system maintenance cost is reduced. This method establishes a complete set of module identification and optimization mechanisms, enhances the stability and scalability of the cross-border service platform, and provides technical support for the continuous optimization and upgrading of the platform. Overall, this method significantly improves the operating efficiency and service quality of the cross-border service platform, providing users with a better cross-border service experience.

[0034] In one embodiment, real-time processing data of the cross-border service platform is obtained, and the real-time processing data is associated with the core service module information to obtain a corresponding real-time service processing group, including: By collecting various data streams generated during the operation of the platform in real time, including user access data, service call data, system operation data, etc., a real-time processing data set is formed. The data collection process is continuous to ensure the real-time and integrity of the data.

[0035] Based on the configured core service module information, the collected real-time processing data is analyzed for redundancy. During the redundancy analysis, the real-time data is matched with the existing service modules to identify duplicate or similar service functions and generate a redundant service relationship list. The redundant service relationship list records information such as duplicate service items, duplication degree, and association relationship between services.

[0036] Conduct in-depth analysis of the service items in the redundant service relationship list, count the call frequency of each service, and analyze the changes in users' actual demand for each service in combination with user behavior data. Through these analyses, the system identifies the optimal service combination method and generates optimization module information. The optimization module information includes service optimization suggestions, service combination solutions, etc.

[0037] The optimization module information is associated and matched with the real-time processing data set to establish a service call relationship diagram to form an initial real-time processing group. The initial real-time processing group shows the optimized service call path and processing flow, providing a basis for subsequent optimization.

[0038] The performance indicators of each service in the initial real-time processing group are calculated, including key indicators such as response time, resource occupancy rate, and service success rate, to obtain initial service indicator data. These indicator data reflect the operating effect of the current service combination.

[0039] Based on the redundant service relationship list generated in the early stage, the initial service indicator data is analyzed to identify performance bottlenecks, adjust service configuration parameters, and generate an optimized service configuration plan. The optimized configuration plan includes specific parameter adjustment values, server resource allocation strategies, and other contents.

[0040] According to the optimized service configuration plan, the initial real-time processing group is optimized and adjusted, including adjusting the server load balancing strategy, optimizing the service call path, merging redundant services, etc., and finally forming a real-time service processing group. This processing group achieves the optimal balance between service performance and resource utilization.

[0041] During the operation of the optimized real-time service processing group, it continuously monitors its operating status and collects performance data to provide a basis for the next round of optimization. The entire optimization process forms a closed loop to ensure continuous improvement of service quality.

[0042] This embodiment realizes the intelligent optimization configuration of service resources by dynamically collecting and analyzing the real-time processing data of the cross-border service platform. In the redundancy analysis stage, duplicate service functions are identified, and dynamic analysis is performed in combination with the service call frequency and changes in user needs, which effectively avoids resource waste and improves the platform operation efficiency. By establishing a service call relationship diagram, service performance indicators are calculated and monitored in real time, performance bottlenecks are discovered in a timely manner, and targeted adjustments are made by optimizing the service configuration plan to ensure the continuous improvement of service quality. A closed-loop optimization mechanism is adopted to continuously collect operation data and make optimization adjustments, so that the platform service performance and resource utilization are optimally balanced, significantly improving the operation stability and service response speed of the cross-border service platform. Through intelligent service combination and load balancing strategies, the scalability of the platform is enhanced, providing users with a better and more efficient cross-border service experience.

[0043] In one embodiment, the security data of the cross-border service platform is obtained, and the security prediction of the core service module information is performed based on the security data to obtain the corresponding security protection trend, including: By collecting and analyzing the security data of the platform, the security prediction of the core service module is realized. In this embodiment, the process of obtaining security data includes multiple key links.

[0044] An automated collection mechanism is set up during the log record collection phase to comprehensively collect all log information generated by the platform operation, including user access logs, operation behavior logs, system operation logs, etc., to form an original security log database. The collection frequency is once every 5 minutes to ensure the real-time and integrity of the data.

[0045] The event classification processing stage uses preset classification rules to classify the original security log data according to the event type. The classification dimensions include: access anomalies, permission violations, data leaks, system vulnerabilities, etc. Each type of event is configured with a corresponding feature identifier, and automatic classification is completed through feature matching to generate a classified security event data set.

[0046] In the time series analysis phase, the sliding time window method is used to perform time series correlation analysis on the classified security event data. The time window is set to 24 hours by default, and the window sliding step is 1 hour. In each time window, the occurrence order and time interval of various security events are recorded to construct a time series security event sequence.

[0047] The frequency distribution statistics phase conducts statistical analysis on the frequency of events in the time series security event sequence. The statistical period is set to 7 days, and the statistical granularity is hourly. By calculating the number of occurrences of each type of event in different time periods, a frequency distribution curve is drawn to form event frequency distribution data.

[0048] In the feature extraction phase, security threat features are extracted based on event frequency distribution data. Feature dimensions include: event frequency, temporal regularity, degree of harm, impact range, etc. The principal component analysis method is used to extract the main feature vectors and construct a security threat feature set.

[0049] In the module matching phase, the security threat feature set is associated with the core service module. The matching rules are based on the module functional attributes, data interaction relationships, business dependencies and other dimensions. By calculating the feature similarity, the relevant modules affected by the threat are determined and a list of security-related modules is generated.

[0050] During the security assessment phase, each module in the security-related module list is scored. The scoring indicators include: threat level, vulnerability level, attack probability, impact range, etc. A weighted scoring method is used to calculate the comprehensive security score and form the module security score data.

[0051] The scenario prediction stage combines the module security score data with the event frequency distribution data to predict potential security threat scenarios. The prediction model is trained based on historical data and uses a time series prediction algorithm to predict the probability of security events in a certain period of time in the future and generate initial protection scenario data.

[0052] In the trend analysis phase, the initial protection scenario data is deeply analyzed based on the security threat feature set. The analysis dimensions include: threat evolution trend, attack mode changes, protection effect evaluation, etc. Through trend fitting and prediction, the security protection trend is obtained, providing a basis for the formulation of subsequent protection measures.

[0053] This embodiment achieves accurate identification and classification of security incidents and improves the accuracy of security data processing by setting up an automated log collection mechanism and multi-dimensional classification rules. The sliding time window method is used for time series analysis to enhance the ability to grasp the time characteristics of security incidents and make security situation assessment more objective. The frequency distribution statistics and principal component analysis methods are combined to extract security threat features, highlight important security features, and reduce data redundancy. By establishing a module matching mechanism and a multi-dimensional scoring system, accurate security assessment of core service modules is achieved, improving the pertinence of security protection.

[0054] In one embodiment, a service planning analysis is performed on the real-time service processing group and the security protection trend to obtain a corresponding initial platform deployment solution, including: When analyzing data traffic for the real-time service processing group, the data traffic in different time periods is sampled and counted to obtain the traffic peak, traffic mean and traffic fluctuation range of each time period. The service load distribution data is formed by normalizing the sampled data. The sampling period includes the peak period of weekdays, the off-peak period of weekdays, the peak period of holidays and the off-peak period of holidays. The sampling frequency is once every 5 minutes and the sampling duration is 30 consecutive days.

[0055] Based on the service load distribution data, we conduct statistical analysis on the resource consumption of each service, including indicators such as CPU usage, memory occupancy, storage space usage, and network bandwidth usage. By establishing a resource evaluation model, we quantify the correspondence between various resource usage indicators and actual business volume, thereby obtaining platform resource demand data. The resource evaluation model uses a linear regression algorithm and uses historical data as a training set for model training.

[0056] In the process of security domain division, services are divided into three security protection areas: core service area, general service area and DMZ area according to the business sensitivity level. The core service area deploys key business systems and important data, and uses double firewall isolation; the general service area deploys general business systems and uses single firewall protection; the DMZ area deploys external service systems and implements security protection through load balancing equipment.

[0057] When optimizing the initial topology, the service nodes in each security zone are grouped according to the business relevance. The groups are interconnected in a star-shaped network, and the groups are interconnected in a mesh-shaped network. The physical deployment location of the service nodes is determined by the minimum spanning tree algorithm to minimize the communication delay between nodes. The bandwidth configuration of the network link is dynamically adjusted according to the service load distribution data.

[0058] In the service capacity evaluation phase, the queuing theory model is used to simulate and analyze the processing capacity of each service node. The service request arrival process is modeled as an M / M / 1 queuing system. By calculating performance indicators such as the system average response time and average queue length, it is evaluated whether the service capacity meets business needs. Evaluation indicators include peak concurrency, average response time, system throughput, etc.

[0059] In the process of developing the load balancing deployment plan, a weighted polling algorithm is used to implement request distribution. The weight value of each service node is dynamically adjusted according to the processing capacity of the node and the current load situation to achieve dynamic load balancing. At the same time, a heartbeat detection mechanism is established between nodes to automatically switch services when a node failure is detected.

[0060] The redundancy calculation is based on the system reliability theory, and the parallel reliability model is used to calculate the overall reliability of the system. For the core business system, the redundant configuration adopts the dual-machine hot standby mode; for the ordinary business system, the N+1 backup mode is adopted; for the non-critical business system, load balancing is adopted as a redundant protection measure.

[0061] Resource scheduling optimization uses a heuristic algorithm, takes system operating costs and service quality as optimization targets, and solves the optimal deployment plan through a genetic algorithm. During the optimization process, multiple evaluation indicators such as hardware resource utilization, network bandwidth utilization, and system response time are considered simultaneously to achieve the overall optimal resource allocation. The fitness function of the genetic algorithm comprehensively considers the weights of various evaluation indicators, and obtains the initial platform deployment plan through multiple generations of iterative optimization.

[0062] This embodiment accurately grasps the service load distribution characteristics and realizes accurate prediction of resource demand by performing multi-dimensional sampling and analysis on data traffic. A comprehensive in-depth defense system is constructed based on a multi-level domain division mechanism based on security sensitivity, combined with protective measures such as double firewall isolation. A star-mesh hybrid networking method is adopted, and the minimum spanning tree algorithm is used to optimize node deployment, which effectively reduces the system communication delay. Capacity assessment is performed through a queuing theory model, combined with a dynamic load balancing mechanism of weighted polling, to ensure the stable operation of the system. In terms of redundant configuration, differentiated backup strategies are adopted according to the importance of the business, which not only ensures the high availability of the core business, but also avoids waste of resources. The multi-objective optimization scheme based on genetic algorithms achieves the optimal balance between system performance and operating costs, and significantly improves the overall service quality and operating efficiency of the platform.

[0063] In one embodiment, abnormality identification is performed on security data to obtain corresponding abnormal information, and a MOSMPC strategy analysis is performed on the security protection trend and abnormal information through a preset multi-objective security model to obtain a corresponding abnormality prediction and processing strategy, including: When identifying anomalies in security data, the feature extraction module extracts data features from the security data. The extracted features include network traffic features such as packet size, packet time interval, packet protocol type, and packet destination port, forming an n-dimensional security feature vector group V={v1, v2, ..., vn}. Feature extraction uses a deep learning method and uses a pre-trained convolutional neural network to extract high-dimensional feature representations.

[0064] Based on the extracted security feature vector group, the improved K-means clustering algorithm is used to perform cluster analysis on the feature vectors. Cluster analysis calculates the Euclidean distance between vectors, divides similar feature vectors into the same category, and obtains k feature cluster center points C={c1, c2, ..., ck}. The number of cluster center points k is determined by the silhouette coefficient to determine the optimal value, and the cluster iteration termination condition is that the change in the center point position is less than the preset threshold or the maximum number of iterations is reached.

[0065] For each eigenvector vi, calculate its distance from all cluster centers and construct a density function based on distance. The density function uses a Gaussian kernel function to map the distance to a density value in the interval [0, 1]. By interpolating the density values ​​of all eigenvectors, a vector density distribution map D is generated. The bandwidth parameter in the density calculation is determined by cross-validation to determine the optimal value.

[0066] The density distribution map D is segmented by adaptive threshold using the OTSU algorithm, and the area with density value lower than the threshold is marked as the abnormal candidate area R. The threshold segmentation adopts the inter-class variance maximization criterion to automatically determine the optimal segmentation threshold. In the abnormal candidate area, the security feature vector is traversed and detected, and the abnormal vector is identified by the preset abnormal judgment rule to generate the abnormal information set E. The abnormal judgment rule is set based on statistical significance test and expert knowledge base.

[0067] The security protection trend data S is decomposed by wavelet, and the trend component, periodic component and random component are extracted to construct the trend feature matrix T. The wavelet basis function selects db4 wavelet, and the decomposition layer number is 3. The abnormal information E is mapped to the high-dimensional feature space through the nonlinear mapping function to generate the abnormal feature matrix A. The mapping function adopts the RBF kernel function form.

[0068] The trend feature matrix T and the anomaly feature matrix A are input into the preset multi-objective security model. The model is based on a multi-objective optimization framework, and the objective function includes multiple indicators such as anomaly detection accuracy, false alarm rate, and protection resource consumption. Through iterative optimization using the MOSMPC algorithm, the Pareto optimal anomaly prediction and processing strategy is obtained while ensuring data privacy. During the MOSMPC iteration process, a secure multi-party computing protocol is used to protect sensitive data, and the iteration termination condition is when the strategy converges or the maximum number of iterations is reached.

[0069] This embodiment can accurately identify abnormal behavior characteristics by performing deep feature extraction and cluster analysis on security data, thereby improving the accuracy and reliability of anomaly detection. The adaptive threshold segmentation mechanism based on the OTSU algorithm realizes the intelligent division of abnormal candidate areas, reduces the false alarm rate, and improves detection efficiency. Wavelet decomposition is used to extract time series features of security protection trends, and combined with nonlinear feature mapping, a complete abnormal feature representation is constructed, which enhances the model's ability to characterize abnormal behaviors. Through the combination of a multi-objective security model and the MOSMPC algorithm, a dynamic balance of multiple security goals is achieved while protecting data privacy, ensuring the optimality and practicality of the anomaly prediction and processing strategy. This method makes full use of the advantages of deep learning and secure multi-party computing, ensures data security while improving anomaly detection performance, and provides reliable technical support for the security protection of cross-border integrated service platforms.

[0070] In one embodiment, the abnormal information and security protection trend are input into a preset multi-objective security model for MOSMPC strategy analysis to obtain a corresponding abnormal prediction and processing strategy, including: The abnormal information and security protection trends are input into the preset multi-objective security model for MOSMPC strategy analysis. The multi-objective security model includes six hierarchical structures: system security layer, application security layer, data security layer, network security layer, physical security layer and policy processing layer.

[0071] After receiving the abnormal information, the system security layer conducts a system-level security situation assessment on the abnormal information based on the preset security indicator system. The security indicator system includes quantitative indicators such as system resource utilization, system response time, system concurrent processing capability, and system fault tolerance. Through the comprehensive evaluation of these indicators, system-level security situation information is generated. This information reflects the overall operating status of the system and potential security risks.

[0072] The application security layer conducts a double analysis of abnormal information. In the business process security analysis phase, the business scenarios where abnormal information occurs are located and evaluated in combination with the business process map; in the key application risk assessment phase, the risk level of the affected application is quantitatively assessed based on the application importance scoring standard. The assessment results form application layer security risk information, including risk level, impact range, repair suggestions, etc.

[0073] The data security layer analyzes the input security protection trend information. In the data flow trajectory analysis, the data flow graph tracking technology is used to model the data transmission path; in the sensitive data distribution analysis, the sensitive data aggregation area is identified based on the preset data classification standards. The analysis results form data security situation information, showing the dynamic change characteristics of the data security situation.

[0074] The network security layer conducts in-depth analysis of security protection trends. Network topology feature analysis extracts network structure features based on graph theory methods; abnormal traffic identification detects deviations based on a preset traffic baseline model. The analysis results form network security status information, reflecting the network security situation.

[0075] The physical security layer analyzes security protection trends from two dimensions. Physical equipment status assessment is based on a quantitative assessment of the equipment health index system; boundary protection analysis evaluates the protection strength based on preset boundary security rules. The assessment results form physical security boundary information.

[0076] The policy processing layer converts the above five types of information into a MOSMPC constraint set function. This function includes a security objective function, a resource constraint function, and a performance constraint function. Based on the distributed computing framework, the constraint set function is decomposed into sub-problems to obtain a preliminary distributed solution. The sub-problem solving process is optimized through a dynamic adjustment algorithm to obtain a local optimal solution.

[0077] The global coordination optimization module optimizes the local optimal solution and the distributed solution as a whole to generate an initial solution. The solution is iteratively optimized through constraint parameters to eventually form an abnormality prediction and processing strategy. The strategy includes specific protection measures, resource scheduling solutions, and emergency response suggestions.

[0078] This embodiment achieves all-round protection of system security, application security, data security, network security and physical security through a multi-level security model structure, and improves the overall security protection capability of the cross-border integrated service platform. Based on the preset security indicator system, a multi-dimensional analysis of abnormal information is performed to effectively identify potential risks in the system operation process and improve the early warning efficiency of potential safety hazards. With the help of data flow graph tracking technology and sensitive data distribution analysis, the data flow trajectory and sensitive information distribution characteristics are accurately grasped, and the accuracy of data security management is enhanced. By constructing the MOSMPC constraint set function, the unified optimization of security goals, resource constraints and performance constraints is achieved, and the scientific nature of the protection strategy is improved. The use of distributed computing and dynamic adjustment algorithms improves the computational efficiency of the strategy optimization process and ensures the timeliness and effectiveness of protective measures. The final abnormal prediction and processing strategy has strong practicality and operability, providing strong support for the safe operation of the cross-border integrated service platform.

[0079] In one embodiment, the initial platform deployment plan and the abnormal prediction and processing strategy are comprehensively analyzed to obtain the corresponding global service optimization strategy, including: When the initial platform deployment plan is decomposed into four layers, the platform deployment plan is decomposed into infrastructure layer, data layer, service layer and application layer. The infrastructure layer contains hardware resource configuration information, network topology information and system component information; the data layer contains data storage architecture information, data processing flow information and data security policy information; the service layer contains service component information, service interaction information and service quality information; the application layer contains business function information, user interface information and application extension information. The multi-layer deployment architecture data is obtained by decomposing these layers.

[0080] In the process of layered mapping, the abnormal prediction and processing strategies are mapped and analyzed according to different levels. The infrastructure layer maps abnormal types such as hardware failure prediction, network interruption prediction, and system crash prediction; the data layer maps abnormal types such as data loss prediction, data pollution prediction, and data leakage prediction; the service layer maps abnormal types such as service timeout prediction, service degradation prediction, and service deadlock prediction; the application layer maps abnormal types such as function abnormality prediction, interface failure prediction, and extension conflict prediction. Through this layered mapping, a multi-dimensional risk assessment matrix is ​​formed.

[0081] In the weighted fusion phase, the hierarchical analysis method is used to determine the weight coefficients of each layer. The weight of the infrastructure layer is 0.3, the weight of the data layer is 0.25, the weight of the service layer is 0.25, and the weight of the application layer is 0.2. These weight coefficients are weighted and calculated with the multi-layer deployment architecture data to generate a fusion service architecture. The fusion service architecture contains the association relationship and coordination mechanism between each layer.

[0082] In the dynamic optimization calculation process, the converged service architecture is analyzed in time series. By establishing a time series optimization model, various optimization parameters are calculated, including resource scheduling parameters, data processing parameters, service response parameters, and application performance parameters. These parameters reflect the time series characteristics of the platform operation process.

[0083] In the iterative optimization phase, the converged service architecture is optimized for multiple rounds based on the timing optimization parameters. Each round of iteration includes three steps: parameter update, architecture adjustment, and performance evaluation. The optimized service solution is obtained through iterative optimization, which improves the platform operation efficiency while ensuring service quality.

[0084] In the constraint verification phase, constraints such as resource utilization not less than 85%, service response time not exceeding 200ms, and data processing delay not exceeding 100ms were set. The optimized service plan was verified and verification result data was generated, including the compliance status of various indicators.

[0085] In the correction processing stage, the non-compliant indicators are optimized according to the verification result data. By adjusting the relevant parameters and optimizing the corresponding mechanism, the optimized service solution meets all the constraints. Finally, a global service optimization strategy is formed, which ensures the stable operation and efficient service of the platform.

[0086] This embodiment achieves refined management of the infrastructure layer, data layer, service layer and application layer by hierarchical decomposition of the initial platform deployment plan and hierarchical mapping with the abnormal prediction and processing strategy, effectively improving the operational reliability of the cross-border comprehensive service platform. Weighted fusion is performed based on the weight coefficients determined by the hierarchical analysis method, so that resources at each level are reasonably allocated and the overall synergy of the platform is significantly enhanced. The timing optimization model is used to dynamically optimize the fusion service architecture, accurately grasp the timing characteristics during the platform operation process, and provide a reliable basis for subsequent optimization. The practicality and feasibility of the optimized service plan are ensured by combining multiple rounds of iterative optimization with constraint verification. The final global service optimization strategy not only ensures the stable operation of the platform, but also achieves a significant improvement in service efficiency, laying a solid foundation for the long-term stable operation of the cross-border comprehensive service platform.

[0087] The present invention also provides a system for constructing a cross-border integrated service platform, which is applied to any of the above-mentioned methods for constructing a cross-border integrated service platform, comprising: The acquisition module is used to obtain the system architecture data and service management information of the cross-border service platform, divide the system architecture data into modules based on the service management information, and obtain the corresponding core service module information; An analysis module is used to obtain real-time processing data of the cross-border service platform, associate the real-time processing data with the core service module information, and obtain the corresponding real-time service processing group; The association module is used to obtain the security data of the cross-border service platform, make security predictions on the core service module information based on the security data, and obtain the corresponding security protection trend; The processing module is used to perform service planning analysis on the real-time service processing group and security protection trend to obtain the corresponding initial platform deployment plan; A control module, which is used to identify abnormalities in security data and obtain corresponding abnormal information; An execution module is used to input abnormal information and security protection trends into a preset multi-objective security model for MOSMPC strategy analysis to obtain a corresponding abnormal prediction and processing strategy; Analysis module, the analysis module is used to comprehensively analyze the initial platform deployment plan and the abnormal prediction and processing strategy to obtain the corresponding global service optimization strategy.

[0088] The system for constructing a cross-border integrated service platform provided by the present invention has the following beneficial effects: By associating real-time processing data with core service modules to form a real-time service processing group, the platform's ability to respond to dynamic business needs is significantly enhanced, ensuring efficient processing of cross-border services. The introduction and predictive analysis of security data enable the platform to have the ability to perceive potential security threats in advance, and provide more accurate risk warnings in combination with security protection trends, providing guarantees for the stable operation of cross-border businesses. By combining abnormal information with multi-objective security models, an abnormal prediction and processing strategy is generated, which can effectively solve the problem of rapid response under dynamic security threats. The comprehensive optimization analysis of the initial platform deployment plan and security strategy enables the platform's resource allocation and service performance to achieve the optimal balance of overall service efficiency while meeting business needs, thereby reducing system operating costs and improving user experience. This global optimization strategy provides important support for building an efficient, secure, and intelligent cross-border service platform.

[0089] It should be noted that technicians in the relevant technical field can clearly understand that for the convenience and conciseness of description, the specific working process of the system and each module described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0090] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for constructing a cross-border integrated service platform, characterized in that: include: Acquire system architecture data and service management information of the cross-border service platform, divide the system architecture data into modules based on the service management information, and obtain corresponding core service module information; Acquire the real-time processing data of the cross-border service platform, associate the real-time processing data with the core service module information, and obtain a corresponding real-time service processing group; Acquire the security data of the cross-border service platform, perform security prediction on the core service module information based on the security data, and obtain corresponding security protection trends; Performing service planning analysis on the real-time service processing group and the security protection trend to obtain a corresponding initial platform deployment plan; Performing abnormality identification on the security data to obtain corresponding abnormality information; Input the abnormal information and the security protection trend into a preset multi-objective security model to perform MOSMPC strategy analysis to obtain a corresponding abnormal prediction and processing strategy; The initial platform deployment plan and the abnormal prediction and processing strategy are comprehensively analyzed to obtain a corresponding global service optimization strategy.

2. The method for constructing a cross-border integrated service platform according to claim 1, characterized in that: The obtaining of system architecture data and service management information of the cross-border service platform, and module division of the system architecture data based on the service management information to obtain corresponding core service module information, includes: Collecting the architecture of the cross-border service platform to obtain original architecture data; Performing redundancy optimization on the original architecture data to obtain the system architecture data; Acquire the service management information of the cross-border service platform, and perform functional relationship analysis on the system architecture data to obtain a module functional relationship diagram; According to the module function relationship diagram, the system architecture data is hierarchically divided to obtain hierarchical module information; Performing performance evaluation on the hierarchical module information to obtain corresponding performance evaluation results; According to the performance evaluation result and the module function relationship diagram, the hierarchical module information is prioritized to obtain corresponding module priority sorting information; A core module analysis is performed based on the module priority sorting information and the service management information to obtain the core service module information.

3. The method for constructing a cross-border integrated service platform according to claim 1, characterized in that: The acquiring of the real-time processing data of the cross-border service platform and associating the real-time processing data with the core service module information to obtain a corresponding real-time service processing group includes: Collecting real-time data streams from the cross-border service platform to obtain the real-time processing data; Performing redundancy analysis on the real-time processing data according to the core service module information to obtain a redundant service relationship list; The redundant service relationship list is subjected to a dynamic analysis of service call frequency and user demand to obtain optimization module information. Performing preliminary correlation matching on the optimization module information and the real-time processing data set to obtain an initial real-time processing group; Calculating the index of the initial real-time processing group to obtain corresponding initial service index data; Optimizing the service configuration of the initial service indicator data according to the redundant service relationship list to obtain a corresponding optimized service configuration; The initial real-time processing group is optimized according to the optimized service configuration to obtain the real-time service processing group.

4. The method for constructing a cross-border integrated service platform according to claim 1, characterized in that: The obtaining of the security data of the cross-border service platform, and performing security prediction on the core service module information according to the security data to obtain a corresponding security protection trend, includes: Collecting log records of the cross-border service platform to obtain original security log data; Performing event classification on the original security log data to obtain classified security event data; Performing time series analysis on the classified security event data to obtain a time series security event sequence; Perform frequency distribution statistics according to the time-series security event sequence to obtain event frequency distribution data; Extracting features from the event frequency distribution data to obtain a security threat feature set; Perform module matching on the core service module information according to the security threat feature set to obtain a security-related module list; Performing a security assessment on the security-related module list to obtain module security score data; Perform scenario prediction on the event frequency distribution data according to the module safety score data to obtain initial protection scenario data; Based on the security threat feature set, trend analysis is performed on the initial protection scenario data to obtain a security protection trend.

5. The method for constructing a cross-border integrated service platform according to claim 1, characterized in that: The performing of service planning analysis on the real-time service processing group and the security protection trend to obtain a corresponding initial platform deployment solution includes: Performing data flow analysis on the real-time service processing group to obtain service load distribution data; Performing resource demand statistics on the service load distribution data to obtain platform resource demand data; Dividing the security protection trend into security domains to obtain security protection area data; Perform topology optimization according to the platform resource demand data and the security protection area data to obtain an initial topology solution; Performing service capacity evaluation on the initial topology structure solution to obtain service capacity evaluation data; Perform load balancing on the initial topology scheme according to the service capacity evaluation data to obtain a load balancing deployment scheme; Performing redundancy calculation on the load balancing deployment scheme to obtain a service redundancy configuration scheme; The service redundancy configuration scheme is optimized for resource scheduling to obtain the initial platform deployment scheme.

6. The method for constructing a cross-border integrated service platform according to claim 1, characterized in that: The abnormality identification is performed on the security data to obtain corresponding abnormality information, and the MOSMPC strategy analysis is performed on the security protection trend and the abnormality information through a preset multi-objective security model to obtain a corresponding abnormality prediction and processing strategy, including: Extracting data features from the security data to obtain a corresponding security feature vector group; Performing cluster analysis on the security feature vector group to obtain corresponding feature cluster center points; Performing density calculation on the security feature vector group according to the feature cluster center point to obtain a corresponding vector density distribution map; Performing threshold segmentation on the vector density distribution map to obtain corresponding abnormal candidate regions; An abnormal traversal detection is performed on the security feature vector group according to the abnormal candidate area to obtain the abnormal information. Performing time series decomposition processing on the security protection trend to obtain a corresponding trend feature matrix; Perform multi-dimensional feature mapping on to obtain the corresponding abnormal feature matrix; The abnormal information and the abnormal feature matrix are input into the preset multi-objective safety model, and MOSMPC iterative processing is performed to obtain the abnormal prediction processing strategy.

7. The method for constructing a cross-border integrated service platform according to claim 1, characterized in that: The abnormal information and the security protection trend are input into a preset multi-objective security model for MOSMPC strategy analysis to obtain a corresponding abnormal prediction and processing strategy, including: Inputting the abnormal information into the multi-objective security model, performing an overall security situation assessment on the abnormal information through the system security layer of the multi-objective security model, and obtaining system-level security situation information; Performing business process security analysis and key application risk assessment on the abnormal information through the application security layer of the multi-objective security model to obtain application layer security risk information; Inputting the security protection trend into the multi-objective security model, performing data flow trajectory and sensitive data distribution analysis on the security protection trend through the data security layer of the multi-objective security model, and obtaining data security situation information; Through the network security layer of the multi-objective security model, network topology feature analysis and abnormal traffic identification are performed on the security protection trend to obtain network security status information; Performing physical device status assessment and boundary protection analysis on the security protection trend through the physical security layer of the multi-objective security model to obtain physical security boundary information; Through the policy processing layer of the multi-objective security model, function constraints are constructed on the system-level security situation information, the application-layer security risk information, the data security situation information, the network security status information and the physical security boundary information to obtain a corresponding MOSMPC constraint condition set function; Solving sub-problems of the MOSMPC constraint condition set function to obtain a corresponding distributed solution; Dynamically adjusting and solving sub-problems of the distributed solution to obtain a local target solution; Performing global coordinated optimization on the local objective solution and the distributed solution to obtain an initial solution; The constraint parameters of the initial solution are iterated to obtain and output the abnormality prediction processing strategy.

8. The method for constructing a cross-border integrated service platform according to claim 1, characterized in that: The initial platform deployment plan and the abnormal prediction and processing strategy are comprehensively analyzed to obtain a corresponding global service optimization strategy, including: Decomposing the initial platform deployment plan hierarchically to obtain corresponding multi-layer deployment architecture data; Performing hierarchical mapping on the anomaly prediction and processing strategy according to the multi-layer deployment architecture data to obtain a corresponding multi-dimensional risk assessment matrix; Performing weighted fusion on the multi-layer deployment architecture data according to the multi-dimensional risk assessment matrix to obtain a corresponding fusion service architecture; Performing dynamic optimization calculation on the fusion service architecture to obtain corresponding timing optimization parameters; Iteratively optimize the fusion service architecture according to the timing optimization parameters to obtain a corresponding optimized service solution; Verifying the constraints of the optimization service solution to obtain corresponding verification result data; The optimized service solution is modified according to the verification result data to obtain a corresponding global service optimization strategy.

9. A system for constructing a cross-border integrated service platform, characterized in that: The method for constructing a cross-border integrated service platform as described in any one of claims 1 to 8 above comprises: A collection module, which is used to obtain system architecture data and service management information of the cross-border service platform, divide the system architecture data into modules based on the service management information, and obtain corresponding core service module information; An analysis module, the analysis module is used to obtain real-time processing data of the cross-border service platform, associate the real-time processing data with the core service module information, and obtain a corresponding real-time service processing group; An association module, the association module is used to obtain the security data of the cross-border service platform, perform security prediction on the core service module information based on the security data, and obtain a corresponding security protection trend; A processing module, the processing module is used to perform service planning analysis on the real-time service processing group and the security protection trend to obtain a corresponding initial platform deployment plan; A control module, the control module is used to identify abnormalities of the security data and obtain corresponding abnormal information; An execution module, wherein the execution module is used to input the abnormal information and the security protection trend into a preset multi-objective security model to perform MOSMPC strategy analysis to obtain a corresponding abnormal prediction and processing strategy; An analysis module is used to comprehensively analyze the initial platform deployment plan and the abnormal prediction and processing strategy to obtain a corresponding global service optimization strategy.

Images