Cross-border e-commerce supervision and customs information integrated management system based on big data

Abstract

The present invention discloses a cross-border e-commerce supervision and customs information integrated management system based on big data, which belongs to the field of cross-border e-commerce supervision technology. The present invention integrates multi-dimensional data from customs, e-commerce platforms and logistics companies, and uses an electronic declaration module to achieve standardized generation and automatic verification of documents, significantly reducing the workload and error rate of manual review, dynamically generating market data maps, and combining the violation prediction model of the analysis and decision-making module to quickly identify potential risks and provide precise support for supervision. Through a multi-level early warning condition rule library and a real-time anomaly detection unit, it can promptly identify abnormal prices, abnormal sales, and other behaviors, generate detailed alarm information and push it to supervisors. At the same time, through data visualization technology, it intuitively presents heat maps and risk reports, helping to quickly identify key targets, improve risk response capabilities, comprehensively optimize the cross-border e-commerce supervision process, and effectively reduce the risk of violations.

Description

Technical Field

[0001] The present invention relates to the field of cross-border e-commerce supervision technology, and in particular to a cross-border e-commerce supervision and customs information integrated management system based on big data. Background Art

[0002] With the rapid development of cross-border e-commerce, the scale and frequency of global commodity circulation have increased significantly, and the traditional customs supervision model faces many challenges.

[0003] The volume of cross-border e-commerce transaction data is huge and comes from diverse sources, covering multiple dimensions such as e-commerce platforms, logistics companies, and customs declaration information. The complexity and dynamism of the data make manual processing difficult to meet the needs of efficient supervision. On the other hand, violations in the cross-border e-commerce field are increasing. At the same time, traditional paper document processing methods are inefficient and have high error rates, making it difficult to adapt to the requirements of modern information management. Summary of the Invention

[0004] The purpose of the present invention is to provide a cross-border e-commerce supervision and customs information integrated management system based on big data to solve the problems raised in the above background technology.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a cross-border e-commerce supervision and customs information integrated management system based on big data, comprising:

[0006] Data acquisition module for:

[0007] Connect to customs data, e-commerce platform data, and third-party logistics company data through interfaces to capture raw data and integrate it into multidimensional data sets;

[0008] Electronic reporting module for:

[0009] Generate standardized electronic declaration document templates and automatically verify and approve declaration document data, digitize historical paper documents and generate a historical database;

[0010] Market monitoring module for:

[0011] Monitor product information on e-commerce platforms, collect product information on e-commerce platforms in real time, generate dynamic market data maps based on product information, and issue early warning signals for abnormal product behavior;

[0012] Analysis and decision-making module for:

[0013] Predict the probability of potential violations, perform semantic analysis on the declared content, identify false information and non-standard descriptions, and generate data visualization reports;

[0014] Information management module for:

[0015] Store multidimensional datasets by subject classification;

[0016] Abnormal monitoring module, used for:

[0017] Set multi-level early warning conditions, identify anomalies in real-time data in multidimensional datasets based on the early warning conditions, and automatically send alarm information to customs management personnel when the early warning conditions are triggered.

[0018] Furthermore, the raw data collected by the data collection module includes customs data, e-commerce platform data, and third-party logistics company data;

[0019] The customs data includes declaration number, declaration time, declarant / company, type of goods, quantity of goods, declared amount, tariff calculation information, goods inspection records, customs clearance status, release records, declaration information of cross-border e-commerce transactions over the years, and customs clearance records;

[0020] The e-commerce platform data includes product name, inventory unit number, category, price, inventory quantity, sales data, seller name, location, registration information, credit score, transaction history, order number, transaction amount, payment time, buyer information, and logistics tracking number;

[0021] The third-party logistics company data includes logistics order number, shipping place, destination, transportation method, transportation time, cargo status and transportation abnormality records.

[0022] Furthermore, the electronic declaration module includes:

[0023] Electronic document generation unit, used for:

[0024] Create an electronic document template based on customs declaration standards, including fixed and dynamic fields. The fixed fields include the declaration number, type of goods, and declared amount, while the dynamic fields include additional tax and fee information and logistics information. A rules engine guides user input, performing real-time verification of the format, scope, and logical consistency of input data to generate an electronic document.

[0025] Automatic document verification unit, used for:

[0026] Build a customs declaration rule library that includes upper and lower limits for declared amounts, and the correspondence between commodity types and tariffs. Verify electronic documents item by item based on the customs declaration rule library, calculate tariffs in real time based on the declared amount and commodity type, and compare them with the user-entered values ​​in the electronic documents. Generate error messages for missing or incorrect data and return them to the user.

[0027] Historical document digitization unit, used for:

[0028] Acquire a document image generated by scanning a paper document with an image acquisition device, perform text recognition on the document image, extract document text data from the document image, automatically fill in an electronic document template based on the document text data, and send the template to the document automatic verification unit for verification;

[0029] The verified electronic documents are stored in the historical database.

[0030] Furthermore, the market monitoring module includes:

[0031] Product information collection unit, used to:

[0032] Establish real-time data connection with e-commerce platforms through data interfaces and capture product information, including product name, inventory unit number, category, price, inventory quantity, sales volume, and seller reputation score; clean the captured product information and build a standardized product information database based on the product information;

[0033] Market map generation unit, used to:

[0034] Build a commodity market model based on commodity categories, price ranges, and sales volume dimensions. Based on the commodity market model, generate a market data map that includes commodity category distribution, sales trends, and price fluctuations. Define the refresh cycle of the dynamic map and update the market data map based on real-time data.

[0035] Furthermore, the refresh cycle of the dynamic graph is defined, including:

[0036] Real-time extraction of the frequency of changes in product category distribution within each preset unit time;

[0037] Extract the frequency of changes in the market data graph of price fluctuations within each preset unit time in real time;

[0038] Obtaining a first-cycle adjustment coefficient using the frequency of change in the commodity category distribution within each preset unit time and the frequency of change in the market data graph of price fluctuations within each preset unit time;

[0039] The first period adjustment coefficient is obtained by the following formula:

[0040] ;

[0041] Among them, L 01 represents the adjustment coefficient of the first cycle; n represents the number of unit time that has passed; f 1i and f 3i They represent the frequency of change in the distribution of commodity categories and the frequency of change in the market data graph of price fluctuations corresponding to the i-th unit time respectively;

[0042] The first cycle adjustment coefficient is compared with a preset first adjustment coefficient threshold, and whether the refresh cycle needs to be redefined is determined according to the comparison result.

[0043] Furthermore, the first cycle adjustment coefficient is compared with a preset first adjustment coefficient threshold, and whether the refresh cycle needs to be redefined is determined according to the comparison result, further comprising:

[0044] Extracting a comparison result between the first period adjustment coefficient and a preset first adjustment coefficient threshold;

[0045] When the comparison result indicates that the first cycle adjustment coefficient does not exceed the preset first adjustment coefficient threshold, it is determined that there is no need to redefine the refresh cycle;

[0046] When the comparison result indicates that the first cycle adjustment coefficient exceeds the preset first adjustment coefficient threshold, the frequency of change of the market data map of commodity category distribution, sales volume trend and price fluctuation in each unit period is retrieved;

[0047] Generate a change frequency matrix using the change frequency of the market data map of commodity category distribution, sales volume trend and price fluctuation within each unit period;

[0048] The change frequency matrix is ​​obtained by the following formula:

[0049] ;

[0050] Among them, A represents the change frequency matrix; f 11 、f 12 and f 13 Respectively represent the frequency of change of the market data map of commodity category distribution, sales trend and price fluctuation corresponding to the first unit time; f 21 、f 22 and f 23 Respectively represent the frequency of change of the market data map of commodity category distribution, sales trend and price fluctuation corresponding to the second unit time; f n1 、f n2 and f n3 The frequencies of changes in the market data graphs representing the commodity category distribution, sales volume trends, and price fluctuations corresponding to the nth unit of time, respectively;

[0051] Obtaining a second period adjustment coefficient using the change frequency matrix;

[0052] The second period adjustment coefficient is obtained by the following formula:

[0053] ;

[0054] Among them, L 02represents the second period adjustment coefficient; n represents the number of unit times that have passed; W represents the weight matrix corresponding to the change frequency matrix; A represents the change frequency matrix; i represents the row index number in the change frequency matrix; α and β represent the first adjustment factor and the second adjustment factor, and the first adjustment factor is obtained by the following formula:

[0055] ;

[0056] Among them, α represents the first adjustment factor; f 1b 、f 2b and f 3b The overall standard deviation of the frequency of changes in the market data graphs representing the commodity category distribution, sales volume trend, and price fluctuation corresponding to n units of time; Indicates the maximum value of the standard deviation corresponding to the change frequency of each row in the change frequency matrix; Indicates the selection of f 1b 、f 2b and f 3b The maximum value in ;

[0057] At the same time, the second adjustment factor is obtained by the following formula:

[0058] ;

[0059] Where β represents the second regulatory factor; f 1b 、f 2b and f 3b The overall standard deviation of the frequency of changes in the market data graphs representing the commodity category distribution, sales volume trend, and price fluctuation corresponding to n units of time; Indicates the minimum value of the standard deviation corresponding to the change frequency of each row in the change frequency matrix; Indicates the selection of f 1b 、f 2b and f 3b The minimum value in ;

[0060] The refresh period is redefined using the first period adjustment coefficient and the second period adjustment coefficient.

[0061] Furthermore, redefining the refresh period using the first period adjustment coefficient and the second period adjustment coefficient includes:

[0062] Retrieve the second cycle adjustment coefficient;

[0063] comparing the second period adjustment coefficient with a preset second adjustment coefficient threshold;

[0064] When the second period adjustment coefficient is not lower than a preset second adjustment coefficient threshold, redefining the refresh period using the first period determination model to obtain an adjusted refresh period;

[0065] The structure of the first cycle determination model is as follows:

[0066] ;

[0067] Among them, T 01 Indicates the refresh cycle after adjustment obtained by the first cycle determination model; T0 indicates the refresh cycle before adjustment; L 01 Indicates the first cycle adjustment coefficient; L 02 Indicates the second cycle adjustment coefficient; L y01 represents the preset first adjustment coefficient threshold; L y02 represents a preset second adjustment coefficient threshold;

[0068] When the second period adjustment coefficient is lower than a preset second adjustment coefficient threshold, redefining the refresh period using the second period determination model to obtain an adjusted refresh period;

[0069] The structure of the second cycle determination model is as follows:

[0070] ;

[0071] Among them, T 02 Indicates the refresh cycle after adjustment obtained by the second cycle determination model; T0 indicates the refresh cycle before adjustment; L 01 Indicates the first cycle adjustment coefficient; L 02 Indicates the second cycle adjustment coefficient; L y01 represents the preset first adjustment coefficient threshold; L y02 Indicates the preset second adjustment coefficient threshold.

[0072] Furthermore, the market monitoring module further includes:

[0073] Abnormal behavior warning unit, used to:

[0074] Building an abnormal behavior rule library, wherein the rules in the abnormal behavior rule library include price anomalies, sales anomalies, inventory anomalies, and seller reputation anomalies;

[0075] Based on the abnormal behavior rule library, the product information obtained in real time is monitored. When products or sellers that meet the abnormal rules are detected, they are marked as potential risk objects and an abnormality report is generated for the potential risk objects. The abnormality report includes the abnormality type, occurrence time and impact scope. Based on the abnormality report, real-time warning signals are pushed to supervisors.

[0076] Furthermore, the analysis and decision-making module includes:

[0077] Violation prediction unit, used to:

[0078] Building a violation prediction model based on historical multidimensional datasets and expert rules, wherein the input parameters of the violation prediction model include declared amount, type of goods, historical customs clearance records, and transaction frequency;

[0079] Obtain real-time multidimensional data sets, input them into the violation prediction model one by one, analyze them, output the violation probability and corresponding high-risk characteristics of each data item, set risk thresholds, mark data exceeding the risk threshold as potential violations, and generate a risk list;

[0080] Semantic analysis unit, used to:

[0081] Build a semantic analysis model for customs declaration scenarios. The model includes keyword extraction, text classification, and anomaly semantic detection modules. It also predefines a standard declaration language template. Based on the semantic analysis model, the model compares electronic declaration documents with the standard declaration language template to detect semantic anomalies in electronic declaration documents.

[0082] A risk report is generated based on the semantic anomaly, wherein the risk report includes the location of the anomaly field and suggestions for modification, and the risk report is returned to the electronic reporting module.

[0083] Furthermore, the analysis and decision-making module further includes:

[0084] Data Visualization Unit for:

[0085] The output results of the violation prediction unit and the semantic analysis unit are classified and summarized by time, region, and cargo type to construct a multidimensional analysis data set, and a risk heat map is generated based on the multidimensional analysis data set.

[0086] Furthermore, the abnormality monitoring module includes:

[0087] The warning rule definition unit is used to:

[0088] Based on historical data analysis and expert experience, a multi-level warning condition rule library is established. The multi-level warning condition rule library includes warning rules for upper and lower limits of declared amounts, warning rules for deviation ranges of goods quantities, and warning rules for abnormal transaction frequency thresholds. The corresponding response method for each warning rule is defined, the rules are mapped to data fields, and the data dimensions to which the rules apply are defined. The data dimensions include time, region, and commodity category.

[0089] Real-time anomaly detection unit for:

[0090] According to the warning rules of the multi-level warning condition rule library, the real-time multidimensional data set is checked one by one, and the data records that meet the warning rule conditions are identified. The data records that meet the warning rule conditions are marked as abnormal risk objects and the type and occurrence time of the triggering warning rule are recorded. Based on the abnormal risk object, alarm information containing the abnormality type, occurrence time, associated warning rules and impact scope is generated.

[0091] Compared with the prior art, the present invention has the following beneficial effects:

[0092] 1. Through efficient data integration and standardized processing in the data acquisition module and electronic declaration module, the present invention can comprehensively collect multi-dimensional data and generate standardized electronic documents. At the same time, combined with automatic verification and real-time validation mechanisms, it significantly reduces manual review time and error rates. Through dynamic market data maps and violation prediction models, it helps regulators quickly identify potential risks and take priority measures for high-risk data, thereby achieving accurate and efficient cross-border e-commerce supervision.

[0093] 2. This invention utilizes the anomaly monitoring module's multi-level warning condition rule base and real-time anomaly detection unit to enable real-time monitoring of cross-border e-commerce declarations, transactions, and logistics. When abnormal behavior is detected, the system promptly generates an alert and sends it to supervisors, ensuring a rapid response to risk events. The analysis and decision-making module, combining semantic analysis with data visualization technology, visually displays potential violations in the form of heat maps and risk reports, assisting supervisors in making informed decisions and improving emergency response capabilities.

[0094] 3. The present invention realizes efficient management and security protection of multi-dimensional data through subject classification storage, retrieval optimization and authority management functions, supports refined authority control and masking of sensitive information, prevents unauthorized data access and information leakage, uses OCR technology to digitize paper documents, and establishes a historical database, providing a solid data foundation for subsequent data analysis and decision support, thereby enhancing the utilization value of data and the sustainable development capability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] Figure 1 This is a schematic diagram of the information integrated management system module of the present invention. DETAILED DESCRIPTION

[0096] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0097] See also Figure 1 , the present invention provides the following technical solutions:

[0098] A comprehensive management system for cross-border e-commerce supervision and customs information based on big data, including:

[0099] Data acquisition module for:

[0100] Connect to customs data, e-commerce platform data, and third-party logistics company data through interfaces to capture raw data and integrate it into multidimensional data sets;

[0101] Electronic reporting module for:

[0102] Generate standardized electronic declaration document templates and automatically verify and approve declaration document data, digitize historical paper documents and generate a historical database;

[0103] Market monitoring module for:

[0104] Monitor product information on e-commerce platforms, collect product information on e-commerce platforms in real time, generate dynamic market data maps based on product information, and issue early warning signals for abnormal product behavior;

[0105] Analysis and decision-making module for:

[0106] Predict the probability of potential violations, perform semantic analysis on the declared content, identify false information and non-standard descriptions, and generate data visualization reports;

[0107] Information management module for:

[0108] Store multidimensional datasets by subject classification;

[0109] Abnormal monitoring module, used for:

[0110] Set multi-level early warning conditions, identify anomalies in real-time data in multidimensional datasets based on the early warning conditions, and automatically send alarm information to customs management personnel when the early warning conditions are triggered.

[0111] The raw data collected by the data collection module includes customs data, e-commerce platform data, and third-party logistics company data;

[0112] The customs data includes declaration number, declaration time, declarant / company, type of goods, quantity of goods, declared amount, tariff calculation information, goods inspection records, customs clearance status, release records, declaration information of cross-border e-commerce transactions over the years, and customs clearance records;

[0113] The e-commerce platform data includes product name, inventory unit number, category, price, inventory quantity, sales data, seller name, location, registration information, credit score, transaction history, order number, transaction amount, payment time, buyer information, and logistics tracking number;

[0114] The third-party logistics company data includes logistics order number, shipping place, destination, transportation method, transportation time, cargo status and transportation abnormality records.

[0115] In the above embodiment, the data acquisition module is the foundation of the entire system. It connects to multiple data sources through interfaces, collects customs data, e-commerce platform data and third-party logistics company data, and integrates them into multidimensional data sets. The diversity of data sources enables the system to obtain multi-dimensional information covering declaration form number, declared amount, cargo quantity, commodity price, transaction records, logistics status, etc., providing comprehensive support for subsequent modules. By connecting with the real-time data interface, the system can capture the latest dynamic information, such as commodity sales, inventory changes, and logistics status updates, ensuring that supervisors can make timely decisions based on the latest data.

[0116] Electronic reporting module, including:

[0117] Electronic document generation unit, used for:

[0118] Create an electronic document template based on customs declaration standards, including fixed and dynamic fields. The fixed fields include the declaration number, type of goods, and declared amount, while the dynamic fields include additional tax and fee information and logistics information. A rules engine guides user input, performing real-time verification of the format, scope, and logical consistency of input data to generate an electronic document.

[0119] Automatic document verification unit, used for:

[0120] Build a customs declaration rule library that includes upper and lower limits for declared amounts, and the correspondence between commodity types and tariffs. Verify electronic documents item by item based on the customs declaration rule library, calculate tariffs in real time based on the declared amount and commodity type, and compare them with the user-entered values ​​in the electronic documents. Generate error messages for missing or incorrect data and return them to the user.

[0121] Historical document digitization unit, used for:

[0122] Acquire a document image generated by scanning a paper document with an image acquisition device, perform text recognition on the document image, extract document text data from the document image, automatically fill in an electronic document template based on the document text data, and send the template to the document automatic verification unit for verification;

[0123] The verified electronic documents are stored in the historical database.

[0124] In the above embodiment, an electronic document template containing fixed fields and dynamic fields is created through pre-defined customs declaration standards, which guides users to fill in documents correctly and verifies the format, range and logical consistency of the input data in real time. By building a customs declaration rule library, the electronic documents submitted by users are automatically checked one by one. It can not only calculate tariffs in real time and compare them with user input values, but also detect missing or incorrect data in the documents and generate specific error prompts, thereby improving the accuracy and standardization of the declared documents. By obtaining the paper document image scanned by the image acquisition device, the unit uses OCR technology to extract the text data in the document and automatically fill it into the electronic document template. The electronic declaration module comprehensively improves the efficiency and accuracy of the declaration process, and provides the possibility of utilizing historical data through digital technology, effectively addressing the common problems of low efficiency and high error rate in cross-border e-commerce declarations.

[0125] Market monitoring module, including:

[0126] Product information collection unit, used to:

[0127] Establish real-time data connection with e-commerce platforms through data interfaces and capture product information, including product name, inventory unit number, category, price, inventory quantity, sales volume, and seller reputation score; clean the captured product information and build a standardized product information database based on the product information;

[0128] Market map generation unit, used to:

[0129] Build a product market model based on product category, price range, and sales volume. Based on the product market model, generate a market data map that includes product category distribution, sales trends, and price fluctuations. Define the refresh cycle of the dynamic map and update the market data map based on real-time data.

[0130] Specifically, the refresh cycle of the dynamic graph is defined as follows:

[0131] Real-time extraction of the frequency of changes in product category distribution within each preset unit time;

[0132] Extract the frequency of changes in the market data graph of price fluctuations within each preset unit time in real time;

[0133] Obtaining a first-cycle adjustment coefficient using the frequency of change in the commodity category distribution within each preset unit time and the frequency of change in the market data graph of price fluctuations within each preset unit time;

[0134] The first period adjustment coefficient is obtained by the following formula:

[0135] ;

[0136] Among them, L 01 represents the adjustment coefficient of the first cycle; n represents the number of unit time that has passed; f 1i and f 3i They represent the frequency of change in the distribution of commodity categories and the frequency of change in the market data graph of price fluctuations corresponding to the i-th unit time respectively;

[0137] The first cycle adjustment coefficient is compared with a preset first adjustment coefficient threshold, and whether the refresh cycle needs to be redefined is determined according to the comparison result.

[0138] The technical effect of the above technical solution is that by extracting the frequency of changes in commodity category distribution and price fluctuation market data maps in real time within each preset unit time, the technical solution can capture the latest information on market dynamics. This real-time performance ensures that the map refresh cycle can keep up with market trends and provide timely and accurate data support. The extracted change frequency is used to calculate the first cycle adjustment coefficient (L 01 This step provides a quantitative assessment of market changes. By comprehensively considering the frequency of changes in both commodity category distribution and price fluctuations, this coefficient can more comprehensively reflect overall market dynamics. By comparing this adjustment coefficient with the preset first adjustment coefficient threshold, it is determined whether the refresh cycle needs to be redefined. This mechanism allows the system to flexibly adjust the refresh cycle based on actual market changes, thereby achieving more refined management. The real-time dynamic map and the dynamically adjusted refresh cycle based on market changes provide decision makers with more timely and accurate market information. This helps them identify market trends more quickly and make more accurate decisions. Furthermore, automated adjustment of the refresh cycle reduces the need for manual intervention, improving the efficiency and accuracy of the decision-making process. By introducing the first cycle adjustment coefficient and a corresponding threshold comparison mechanism, this technical solution enables the system to automatically adjust its refresh cycle based on market changes. This adaptive capability enables the system to better cope with complex and volatile market environments and maintain its stability and effectiveness. By dynamically adjusting the refresh cycle, this technical solution ensures information timeliness while avoiding unnecessary frequent updates. This helps optimize system resource utilization and reduces unnecessary computing and storage overhead.

[0139] In summary, this technical solution achieves accurate capture and timely response to market dynamics through real-time market data extraction, quantitative assessment of market changes, and dynamic adjustment of refresh cycles. This not only improves decision-making efficiency and accuracy, but also enhances the system's adaptability and resource utilization efficiency.

[0140] Specifically, the first cycle adjustment coefficient is compared with a preset first adjustment coefficient threshold, and whether the refresh cycle needs to be redefined is determined according to the comparison result, further comprising:

[0141] Extracting a comparison result between the first period adjustment coefficient and a preset first adjustment coefficient threshold;

[0142] When the comparison result indicates that the first cycle adjustment coefficient does not exceed the preset first adjustment coefficient threshold, it is determined that there is no need to redefine the refresh cycle;

[0143] When the comparison result indicates that the first cycle adjustment coefficient exceeds the preset first adjustment coefficient threshold, the frequency of change of the market data map of commodity category distribution, sales volume trend and price fluctuation in each unit period is retrieved;

[0144] Generate a change frequency matrix using the change frequency of the market data map of commodity category distribution, sales volume trend and price fluctuation within each unit period;

[0145] The frequency matrix is ​​obtained by the following formula:

[0146] ;

[0147] Among them, A represents the change frequency matrix; f 11 、f 12 and f 13 Respectively represent the frequency of change of the market data map of commodity category distribution, sales trend and price fluctuation corresponding to the first unit time; f 21 、f 22 and f 23 Respectively represent the frequency of change of the market data map of commodity category distribution, sales trend and price fluctuation corresponding to the second unit time; f n1 、f n2 and f n3 The frequencies of changes in the market data graphs representing the commodity category distribution, sales volume trends, and price fluctuations corresponding to the nth unit of time, respectively;

[0148] Obtaining a second period adjustment coefficient using the change frequency matrix;

[0149] The second period adjustment coefficient is obtained by the following formula:

[0150] ;

[0151] Among them, L 02represents the second period adjustment coefficient; n represents the number of unit times that have passed; W represents the weight matrix corresponding to the change frequency matrix; A represents the change frequency matrix; i represents the row index number in the change frequency matrix; α and β represent the first adjustment factor and the second adjustment factor, and the first adjustment factor is obtained by the following formula:

[0152] ;

[0153] Among them, α represents the first adjustment factor; f 1b 、f 2b and f 3b The overall standard deviation of the frequency of changes in the market data graphs representing the commodity category distribution, sales volume trend, and price fluctuation corresponding to n units of time; Indicates the maximum value of the standard deviation corresponding to the change frequency of each row in the change frequency matrix; Indicates the selection of f 1b 、f 2b and f 3b The maximum value in ;

[0154] At the same time, the second adjustment factor is obtained by the following formula:

[0155] ;

[0156] Where β represents the second regulatory factor; f 1b 、f 2b and f 3b The overall standard deviation of the frequency of changes in the market data graphs representing the commodity category distribution, sales volume trend, and price fluctuation corresponding to n units of time; Indicates the minimum value of the standard deviation corresponding to the change frequency of each row in the change frequency matrix; Indicates the selection of f 1b 、f 2b and f 3b The minimum value in ;

[0157] The refresh period is redefined using the first period adjustment coefficient and the second period adjustment coefficient.

[0158] The technical solution described above has the following technical effects: By comparing the first-cycle adjustment coefficient with a preset first-cycle adjustment coefficient threshold, the solution intelligently determines whether the refresh cycle needs to be redefined. This mechanism ensures that the map refresh cycle can be dynamically adjusted based on market changes, thereby providing more accurate and timely market information. When redefining the refresh cycle, the solution not only considers the frequency of changes in the market data map, including product category distribution and price fluctuations, but also incorporates the important dimension of sales trends. By integrating and analyzing data from these three dimensions, a more comprehensive assessment of market changes can be achieved, improving the comprehensiveness and accuracy of decision-making. By redefining the refresh cycle using the change frequency matrix and the second-cycle adjustment coefficient, the solution achieves refined adjustment based on market changes. This adjustment mechanism ensures that the map refresh cycle is neither too frequent nor too late, thereby ensuring information timeliness while optimizing system resource utilization. By introducing the first and second adjustment factors, the solution adaptively adjusts the second-cycle adjustment coefficient based on the overall standard deviation of the market data map's frequency of change and the maximum and minimum standard deviations of the frequency of change for each row. This adaptive adjustment capability enables the system to better cope with complex and volatile market environments, improving system stability and robustness. This technical solution reduces the need for manual intervention by automating steps such as extracting data, calculating adjustment coefficients, generating a change frequency matrix, and redefining refresh cycles. This helps improve the efficiency of the decision-making process, enabling decision makers to react more quickly and seize market opportunities. By encapsulating these execution steps in the backend, front-end users can intuitively see the changing trends of market data and the adjustments to the refresh cycle, making it easier to make decisions.

[0159] In summary, this technical solution achieves precise capture and timely response to market dynamics through its advantages in multi-dimensional data fusion analysis, refined regulation, adaptive adjustment capabilities, and improved decision-making efficiency. This not only improves the accuracy and efficiency of decision-making, but also enhances the system's adaptability and resource utilization efficiency.

[0160] Specifically, redefining the refresh period by using the first period adjustment coefficient and the second period adjustment coefficient includes:

[0161] Retrieve the second cycle adjustment coefficient;

[0162] comparing the second period adjustment coefficient with a preset second adjustment coefficient threshold;

[0163] When the second period adjustment coefficient is not lower than a preset second adjustment coefficient threshold, redefining the refresh period using the first period determination model to obtain an adjusted refresh period;

[0164] The structure of the first cycle determination model is as follows:

[0165] ;

[0166] Among them, T 01 Indicates the refresh cycle after adjustment obtained by the first cycle determination model; T0 indicates the refresh cycle before adjustment; L 01 Indicates the first cycle adjustment coefficient; L 02 Indicates the second cycle adjustment coefficient; L y01 represents the preset first adjustment coefficient threshold; L y02 represents a preset second adjustment coefficient threshold;

[0167] When the second period adjustment coefficient is lower than a preset second adjustment coefficient threshold, redefining the refresh period using the second period determination model to obtain an adjusted refresh period;

[0168] The structure of the second cycle determination model is as follows:

[0169] ;

[0170] Among them, T 02 Indicates the refresh cycle after adjustment obtained by the second cycle determination model; T0 indicates the refresh cycle before adjustment; L 01 Indicates the first cycle adjustment coefficient; L 02 Indicates the second cycle adjustment coefficient; L y01 represents the preset first adjustment coefficient threshold; L y02 Indicates the preset second adjustment coefficient threshold.

[0171] The technical effect of the above-mentioned technical solution is that, by introducing a first-cycle adjustment coefficient and a second-cycle adjustment coefficient, it achieves dual adjustment of the refresh cycle. This mechanism enables the system to more comprehensively consider market changes, including product category distribution, sales trends, price fluctuations, and other dimensions, thereby more accurately adjusting the refresh cycle to adapt to market dynamics. This dual adjustment mechanism enhances the system's adaptability and flexibility. Using preset first and second adjustment coefficient thresholds, the technical solution automatically determines whether the refresh cycle needs to be adjusted and which model to use for this adjustment. This intelligent decision-making process reduces manual intervention, improving decision-making efficiency and accuracy. It also enables the system to more quickly respond to market changes and seize business opportunities. The structural design of the first and second cycle determination models allows for adjustments based on actual needs. This flexibility enables the system to adapt to different market environments and business needs, thereby providing more personalized services. For example, the parameters in the model can be adjusted based on the market fluctuation characteristics of a specific industry to obtain a more accurate refresh cycle. By intelligently adjusting the refresh cycle, the technical solution avoids unnecessary frequent updates, thereby optimizing system resource utilization. This helps reduce operating costs and improve cost-effectiveness. At the same time, accurate refresh cycles also help improve data timeliness and accuracy, providing users with more reliable market information. For users of the system, the intelligent refresh cycle adjustment mechanism means they can obtain more timely and accurate market data. This will help them make more informed decisions and improve business efficiency. Furthermore, the system's flexibility and customizability will enhance user satisfaction and loyalty.

[0172] Furthermore, this technical solution enables flexible adjustment of the refresh cycle by introducing a first-cycle adjustment coefficient and a second-cycle adjustment coefficient. Different cycle determination models are selected based on the comparison of these coefficients with preset thresholds. This flexibility enables the system to set a more accurate refresh cycle based on varying market environments and business needs. The structural design of the first-cycle and second-cycle determination models takes into account multiple factors, including the pre-adjusted refresh cycle, the first-cycle adjustment coefficient, the second-cycle adjustment coefficient, and the preset adjustment coefficient threshold. The combined effect of these factors enables the system to more comprehensively assess market changes and accurately set the refresh cycle accordingly. This technical solution sets the refresh cycle through an automated decision-making process and model calculations, reducing the need for human judgment and intervention. This helps reduce errors caused by human factors and improves the accuracy and reliability of cycle setting. This technical solution extracts and analyzes market data in real time, including key information such as product category distribution, sales trends, and price fluctuations. By dynamically adjusting the refresh cycle, the system can respond more quickly to market changes, ensuring that the provided market data is highly timely and accurate. Accurate refresh cycle setting helps users obtain the latest market data in a timely manner, enabling them to make more informed decisions. At the same time, automated decision-making processes and model calculations also improve decision-making efficiency, enabling users to seize business opportunities more quickly.

[0173] In summary, this technical solution has achieved significant results in terms of cycle setting accuracy. By introducing methods such as a dual adjustment mechanism, an automated decision-making process, and model calculation, the system can more flexibly and accurately set refresh cycles, reduce human error, adapt to market changes, and improve decision-making efficiency. These advantages make this technical solution more valuable and meaningful in practical applications. Furthermore, by introducing innovative features such as a dual adjustment mechanism and an intelligent decision-making process, this technical solution achieves intelligent adjustment of the refresh cycle. This not only improves the system's adaptability and flexibility, but also optimizes resource utilization and cost-effectiveness, enhancing user experience and satisfaction.

[0174] Abnormal behavior warning unit, used to:

[0175] Building an abnormal behavior rule library, wherein the rules in the abnormal behavior rule library include price anomalies, sales anomalies, inventory anomalies, and seller reputation anomalies;

[0176] Based on the abnormal behavior rule library, the product information obtained in real time is monitored. When products or sellers that meet the abnormal rules are detected, they are marked as potential risk objects and an abnormality report is generated for the potential risk objects. The abnormality report includes the abnormality type, occurrence time and impact scope. Based on the abnormality report, real-time warning signals are pushed to supervisors.

[0177] In the above embodiment, real-time data docking is established with mainstream e-commerce platforms through data interfaces, key information such as product names, prices, sales, inventory, etc. is captured, and a structured product information database is constructed, which lays the foundation for subsequent analysis. Based on the collected product information, a product market model is constructed, and a dynamic market data map is generated to visualize market dynamic changes, helping regulators identify hot-selling products in the market and determine whether there are abnormal price wars or false promotions through price fluctuation analysis. By building an abnormal behavior rule library, the system can continuously monitor real-time product information. When abnormal behavior is detected, the system will generate an abnormal report and push an early warning signal. The market monitoring module provides effective support for cross-border e-commerce supervision through accurate data collection, comprehensive market analysis and real-time abnormal warning, helping regulators quickly identify potential risks in the market and thus maintain market order.

[0178] Analysis and decision-making module, including:

[0179] Violation prediction unit, used to:

[0180] Building a violation prediction model based on historical multidimensional datasets and expert rules, wherein the input parameters of the violation prediction model include declared amount, type of goods, historical customs clearance records, and transaction frequency;

[0181] Obtain real-time multidimensional data sets, input them into the violation prediction model one by one, analyze them, output the violation probability and corresponding high-risk characteristics of each data item, set risk thresholds, mark data exceeding the risk threshold as potential violations, and generate a risk list;

[0182] Semantic analysis unit, used to:

[0183] Build a semantic analysis model for customs declaration scenarios. The model includes keyword extraction, text classification, and anomaly semantic detection modules. It also predefines a standard declaration language template. Based on the semantic analysis model, the model compares electronic declaration documents with the standard declaration language template to detect semantic anomalies in electronic declaration documents.

[0184] Generate a risk report based on semantic anomalies, including the location of the anomaly field and suggestions for modification, and return the risk report to the electronic reporting module;

[0185] Data Visualization Unit for:

[0186] The output results of the violation prediction unit and the semantic analysis unit are classified and summarized by time, region, and cargo type to construct a multidimensional analysis data set, and a risk heat map is generated based on the multidimensional analysis data set.

[0187] In the above embodiment, by constructing a historical multidimensional dataset and a violation prediction model based on expert rules, potential violations are identified. After comparative analysis with historical data, the probability of violation for each declared data item is predicted. High-risk data is marked as potential violations based on risk thresholds, and a risk list is generated to help supervisors quickly identify key targets. For example, when a declared amount is significantly lower than the average for similar products and occurs frequently, the system will issue an alert. This function not only improves the accuracy of supervision but also significantly reduces the workload of manual investigation.

[0188] In the above example, a semantic analysis model tailored to customs declaration scenarios is constructed to compare electronic declaration documents with standard declaration language templates, detecting semantic anomalies in the text. For example, the system can identify inappropriate descriptions in declaration documents (such as "special goods" instead of specific categories) and generate a risk report. This risk report details the location of inappropriate fields and provides correction suggestions, providing a basis for improvement for declarants and regulators. By identifying false information and non-standard descriptions, it effectively reduces the likelihood of information fraud.

[0189] In the above example, data is aggregated by time, region, and cargo type, and a risk heat map is generated based on the multidimensional analysis dataset. For example, regulators can use the heat map to observe the concentration of recent high-risk declarations in a particular region and deploy targeted inspections. This visualization makes complex data analysis results easier to understand and apply, significantly improving decision-making efficiency.

[0190] Abnormal monitoring module, including:

[0191] The warning rule definition unit is used to:

[0192] Based on historical data analysis and expert experience, a multi-level warning condition rule library is established. The multi-level warning condition rule library includes warning rules for upper and lower limits of declared amounts, warning rules for deviation ranges of goods quantities, and warning rules for abnormal transaction frequency thresholds. The corresponding response method for each warning rule is defined, the rules are mapped to data fields, and the data dimensions to which the rules apply are defined. The data dimensions include time, region, and commodity category.

[0193] Real-time anomaly detection unit for:

[0194] According to the warning rules of the multi-level warning condition rule library, the real-time multidimensional data set is checked one by one, and the data records that meet the warning rule conditions are identified. The data records that meet the warning rule conditions are marked as abnormal risk objects and the type and occurrence time of the triggering warning rule are recorded. Based on the abnormal risk object, alarm information containing the abnormality type, occurrence time, associated warning rules and impact scope is generated.

[0195] In the above-described embodiment, by combining historical data analysis with expert experience to establish a multi-level warning condition rule base, more stringent warning conditions can be set for specific high-risk commodity categories, ensuring accurate monitoring of key areas. Multidimensional data sets are then checked against each warning rule, and abnormal data is quickly identified. For example, when the declared value of a particular commodity falls far below the historical average, the system immediately generates an alert and marks it as a potential risk item. Simultaneously, the detection unit records the rule type and time of the triggering alert and generates an alert report containing the anomaly type, impact area, and detailed records.

[0196] When a warning condition is triggered, the system automatically sends a real-time alert to regulators, reminding them to pay attention to potential risk targets. The anomaly monitoring module, through scientific rule definitions and efficient real-time monitoring, provides regulators with a powerful risk identification tool. The flexibility and real-time nature of this warning mechanism will help comprehensively enhance the sensitivity and responsiveness of cross-border e-commerce supervision, minimizing the risk of violations.

[0197] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A cross-border e-commerce supervision and customs information integrated management system based on big data, characterized by: include: Data acquisition module for: Connect to customs data, e-commerce platform data, and third-party logistics company data through interfaces to capture raw data and integrate it into multidimensional data sets; Electronic reporting module for: Generate standardized electronic declaration document templates and automatically verify and approve declaration document data, digitize historical paper documents and generate a historical database; The electronic declaration module includes: Electronic document generation unit, used for: Create an electronic document template based on customs declaration standards, including fixed and dynamic fields. The fixed fields include the declaration number, type of goods, and declared amount, while the dynamic fields include additional tax and fee information and logistics information. A rules engine guides user input, performing real-time verification of the format, scope, and logical consistency of input data to generate an electronic document. Automatic document verification unit, used for: Build a customs declaration rule library that includes upper and lower limits for declared amounts, and the correspondence between commodity types and tariffs. Verify electronic documents item by item based on the customs declaration rule library, calculate tariffs in real time based on the declared amount and commodity type, and compare them with the user-entered values ​​in the electronic documents. Generate error messages for missing or incorrect data and return them to the user. Historical document digitization unit, used for: Acquire a document image generated by scanning a paper document with an image acquisition device, perform text recognition on the document image, extract document text data from the document image, automatically fill in an electronic document template based on the document text data, and send the template to the document automatic verification unit for verification; Store the verified electronic documents into the historical database; Market monitoring module for: Monitor product information on e-commerce platforms, collect product information on e-commerce platforms in real time, generate dynamic market data maps based on product information, and issue early warning signals for abnormal product behavior; Wherein, the market monitoring module includes: Product information collection unit, used to: Establish real-time data connection with e-commerce platforms through data interfaces and capture product information, including product name, inventory unit number, category, price, inventory quantity, sales volume, and seller reputation score; clean the captured product information and build a standardized product information database based on the product information; Market map generation unit, used to: Build a product market model based on product category, price range, and sales volume. Based on the product market model, generate a market data map that includes product category distribution, sales trends, and price fluctuations. Define the refresh cycle of the dynamic map and update the market data map based on real-time data. The refresh cycle of the dynamic graph is defined, including: Real-time extraction of the frequency of changes in product category distribution within each preset unit time; Extract the frequency of changes in the market data graph of price fluctuations within each preset unit time in real time; Obtaining a first-cycle adjustment coefficient using the frequency of change in the commodity category distribution within each preset unit time and the frequency of change in the market data graph of price fluctuations within each preset unit time; The first period adjustment coefficient is obtained by the following formula: ； Among them, L 01 represents the adjustment coefficient of the first cycle; n represents the number of unit time that has passed; f 1i and f 3i They represent the frequency of change in the distribution of commodity categories and the frequency of change in the market data graph of price fluctuations corresponding to the i-th unit time respectively; Compare the first cycle adjustment coefficient with a preset first adjustment coefficient threshold, and determine whether the refresh cycle needs to be redefined based on the comparison result. Analysis and decision-making module for: Predict the probability of potential violations, perform semantic analysis on the declared content, identify false information and non-standard descriptions, and generate data visualization reports; Information management module for: Store multidimensional datasets by subject classification; Abnormal monitoring module, used for: Set multi-level early warning conditions, identify anomalies in real-time data in multidimensional datasets based on the early warning conditions, and automatically send alarm information to customs management personnel when the early warning conditions are triggered.

2. The cross-border e-commerce supervision and customs information integrated management system based on big data according to claim 1 is characterized in that: The raw data collected by the data collection module includes customs data, e-commerce platform data, and third-party logistics company data; The customs data includes declaration number, declaration time, declarant / company, type of goods, quantity of goods, declared amount, tariff calculation information, goods inspection records, customs clearance status, release records, declaration information of cross-border e-commerce transactions over the years, and customs clearance records; The e-commerce platform data includes product name, inventory unit number, category, price, inventory quantity, sales data, seller name, location, registration information, credit score, transaction history, order number, transaction amount, payment time, buyer information, and logistics tracking number; The third-party logistics company data includes logistics order number, shipping place, destination, transportation method, transportation time, cargo status and transportation abnormality records.

3. The cross-border e-commerce supervision and customs information integrated management system based on big data according to claim 1 is characterized in that: Comparing the first cycle adjustment coefficient with a preset first adjustment coefficient threshold, and determining whether the refresh cycle needs to be redefined according to the comparison result, further comprising: Extracting a comparison result between the first period adjustment coefficient and a preset first adjustment coefficient threshold; When the comparison result indicates that the first cycle adjustment coefficient does not exceed the preset first adjustment coefficient threshold, it is determined that there is no need to redefine the refresh cycle; When the comparison result indicates that the first cycle adjustment coefficient exceeds the preset first adjustment coefficient threshold, the frequency of change of the market data map of commodity category distribution, sales volume trend and price fluctuation in each unit period is retrieved; Generate a change frequency matrix using the change frequency of the market data map of commodity category distribution, sales volume trend and price fluctuation within each unit period; The change frequency matrix is ​​obtained by the following formula: ； Among them, A represents the change frequency matrix; f 11 、f 12 and f 13 Respectively represent the frequency of change of the market data map of commodity category distribution, sales trend and price fluctuation corresponding to the first unit time; f 21 、f 22 and f 23 Respectively represent the frequency of change of the market data map of commodity category distribution, sales trend and price fluctuation corresponding to the second unit time; f n1 、f n2 and f n3 The frequencies of changes in the market data graphs representing the commodity category distribution, sales volume trends, and price fluctuations corresponding to the nth unit of time, respectively; Obtaining a second period adjustment coefficient using the change frequency matrix; The second period adjustment coefficient is obtained by the following formula: ； Among them, L 02 represents the second period adjustment coefficient; n represents the number of unit times that have passed; W represents the weight matrix corresponding to the change frequency matrix; A represents the change frequency matrix; i represents the row index number in the change frequency matrix; α and β represent the first adjustment factor and the second adjustment factor, and the first adjustment factor is obtained by the following formula: ； Among them, α represents the first adjustment factor; f 1b 、f 2b and f 3b The overall standard deviation of the frequency of changes in the market data graphs representing the commodity category distribution, sales volume trend, and price fluctuation corresponding to n units of time; Indicates the maximum value of the standard deviation corresponding to the change frequency of each row in the change frequency matrix; Indicates the selection of f 1b 、f 2b and f 3b The maximum value in ; At the same time, the second adjustment factor is obtained by the following formula: ； Where β represents the second regulatory factor; f 1b 、f 2b and f 3b The overall standard deviation of the frequency of changes in the market data graphs representing the commodity category distribution, sales volume trend, and price fluctuation corresponding to n units of time; Indicates the minimum value of the standard deviation corresponding to the change frequency of each row in the change frequency matrix; Indicates the selection of f 1b 、f 2b and f 3b The minimum value in ; The refresh period is redefined using the first period adjustment coefficient and the second period adjustment coefficient.

4. The cross-border e-commerce supervision and customs information integrated management system based on big data according to claim 3 is characterized in that: Redefining the refresh period by using the first period adjustment coefficient and the second period adjustment coefficient includes: Retrieve the second cycle adjustment coefficient; comparing the second period adjustment coefficient with a preset second adjustment coefficient threshold; When the second period adjustment coefficient is not lower than a preset second adjustment coefficient threshold, redefining the refresh period using the first period determination model to obtain an adjusted refresh period; The structure of the first cycle determination model is as follows: ； Among them, T 01 Indicates the refresh cycle after adjustment obtained by the first cycle determination model; T0 indicates the refresh cycle before adjustment; L 01 Indicates the first cycle adjustment coefficient; L 02 Indicates the second cycle adjustment coefficient; L y01 represents the preset first adjustment coefficient threshold; L y02 represents a preset second adjustment coefficient threshold; When the second period adjustment coefficient is lower than a preset second adjustment coefficient threshold, redefining the refresh period using the second period determination model to obtain an adjusted refresh period; The structure of the second cycle determination model is as follows: ； Among them, T 02 Indicates the refresh cycle after adjustment obtained by the second cycle determination model; T0 indicates the refresh cycle before adjustment; L 01 Indicates the first cycle adjustment coefficient; L 02 Indicates the second cycle adjustment coefficient; L y01 represents the preset first adjustment coefficient threshold; L y02 Indicates the preset second adjustment coefficient threshold.

5. The cross-border e-commerce supervision and customs information integrated management system based on big data according to claim 1, characterized in that: The market monitoring module further includes: Abnormal behavior warning unit, used to: Building an abnormal behavior rule library, wherein the rules in the abnormal behavior rule library include price anomalies, sales anomalies, inventory anomalies, and seller reputation anomalies; Based on the abnormal behavior rule library, the product information obtained in real time is monitored. When products or sellers that meet the abnormal rules are detected, they are marked as potential risk objects and an abnormality report is generated for the potential risk objects. The abnormality report includes the abnormality type, occurrence time and impact scope. Based on the abnormality report, real-time warning signals are pushed to supervisors.

6. The cross-border e-commerce supervision and customs information integrated management system based on big data according to claim 1, characterized in that: The analysis and decision-making module includes: Violation prediction unit, used to: Building a violation prediction model based on historical multidimensional datasets and expert rules, wherein the input parameters of the violation prediction model include declared amount, type of goods, historical customs clearance records, and transaction frequency; Obtain real-time multidimensional data sets, input them into the violation prediction model one by one, analyze them, output the violation probability and corresponding high-risk characteristics of each data item, set risk thresholds, mark data exceeding the risk threshold as potential violations, and generate a risk list; Semantic analysis unit, used to: Build a semantic analysis model for customs declaration scenarios. The model includes keyword extraction, text classification, and anomaly semantic detection modules. It also predefines a standard declaration language template. Based on the semantic analysis model, the model compares electronic declaration documents with the standard declaration language template to detect semantic anomalies in electronic declaration documents. Generate a risk report based on semantic anomalies, including the location of the anomaly field and suggestions for modification, and return the risk report to the electronic reporting module; Data Visualization Unit for: The output results of the violation prediction unit and the semantic analysis unit are classified and summarized by time, region, and cargo type to construct a multidimensional analysis data set, and a risk heat map is generated based on the multidimensional analysis data set.

7. The cross-border e-commerce supervision and customs information integrated management system based on big data according to claim 1, characterized in that: The abnormality monitoring module includes: Warning rule definition unit, used to: Based on historical data analysis and expert experience, a multi-level warning condition rule library is established. The multi-level warning condition rule library includes warning rules for upper and lower limits of declared amounts, warning rules for deviation ranges of goods quantities, and warning rules for abnormal transaction frequency thresholds. The corresponding response method for each warning rule is defined, the rules are mapped to data fields, and the data dimensions to which the rules apply are defined. The data dimensions include time, region, and commodity category. Real-time anomaly detection unit for: According to the warning rules of the multi-level warning condition rule library, the real-time multidimensional data set is checked one by one, and the data records that meet the warning rule conditions are identified. The data records that meet the warning rule conditions are marked as abnormal risk objects and the type and occurrence time of the triggering warning rule are recorded. Based on the abnormal risk object, alarm information containing the abnormality type, occurrence time, associated warning rules and impact scope is generated.

Images