Customs comprehensive information intelligent supervision system based on big data and artificial intelligence
The comprehensive information intelligent supervision system of customs through big data and artificial intelligence technology has solved the problem that the customs supervision system cannot achieve all-round monitoring and early warning, improved the effectiveness and efficiency of customs supervision, and realized the automation and intelligence of customs business.
Patent Information
- Application Number
- CN202411415735.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-11
- Publication Date
- 2024-11-08
AI Technical Summary
The existing customs supervision system cannot achieve comprehensive monitoring and early warning of customs business, resulting in poor supervision and inefficient efficiency.
Adopt a comprehensive customs information intelligent supervision system based on big data and artificial intelligence, including data collection, processing, model training, evaluation optimization, supervision analysis and intelligent control modules. By building a comprehensive customs information intelligent supervision model, it realizes the automation of customs business and Intelligent supervision.
It has achieved all-round monitoring and early warning of customs business, improved the effectiveness and efficiency of customs supervision, and ensured the automation and intelligence of customs business processes.
Smart Images

Figure CN118917810A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of customs supervision technology, and specifically to a customs comprehensive information intelligent supervision system based on big data and artificial intelligence. Background Art
[0002] As global trade becomes increasingly frequent and complex, Customs, as an important department of national border management, shoulders the important responsibility of ensuring national security and maintaining trade order.
[0003] The Chinese patent with publication number CN114897481A discloses a blockchain-based customs logistics intelligent monitoring method, which includes the following steps: determine the various participants of the alliance, each participant establishes a logistics supervision platform based on blockchain technology, and writes the information of each participant into the logistics supervision platform; each participant updates and organizes the logistics information in real time and stores it on the chain; after the participant updates the on-chain logistics information, the logistics supervision platform compares and monitors the logistics information with the logistics information of the previous link; the logistics supervision platform monitors the timeliness of each link of logistics and issues an alarm when the link timeliness exceeds the preset threshold; the logistics supervision platform packages the logistics information of the logistics order after the logistics order is completed and stores it in the file; it can effectively improve the data authenticity of the logistics information, effectively avoid the phenomenon of goods loss or goods substitution, and realize the query and traceability of logistics data; however, the patent has the following defects: The existing system cannot conduct intelligent supervision on comprehensive customs information, and cannot realize all-round monitoring and early warning of customs business, resulting in poor customs supervision effect and low customs supervision efficiency. Summary of the invention
[0004] The purpose of the present invention is to provide a customs comprehensive information intelligent supervision system based on big data and artificial intelligence, which can realize all-round monitoring and early warning of customs business, realize the automation and intelligence of customs business processes, make customs supervision more effective, improve customs supervision efficiency, and solve the problems raised in the above-mentioned background technology.
[0005] To achieve the above object, the present invention provides the following technical solutions: The customs comprehensive information intelligent supervision system based on big data and artificial intelligence includes: Data collection module, used to collect comprehensive real-time customs data based on big data; Data processing module, used to process comprehensive real-time customs data based on big data; Model training module, used to build an AI-based customs comprehensive information intelligent supervision model; Evaluation and optimization module, used to evaluate and optimize the customs comprehensive information intelligent supervision model; The supervision and analysis module is used to conduct supervision and analysis on the comprehensive characteristic data of customs and determine the intelligent supervision and analysis results of comprehensive customs information based on artificial intelligence; Intelligent management and control module is used to carry out intelligent supervision of comprehensive customs information.
[0006] Preferably, the data acquisition module includes: The luggage collection unit is used to monitor the user's luggage and mail items in real time and obtain the user's luggage data; Import and export cargo collection unit, used to monitor import and export cargo in real time and obtain import and export cargo data; The inbound and outbound means collection unit is used to monitor the inbound and outbound means of transport in real time and obtain the inbound and outbound means of transport data; Among them, comprehensive real-time customs data based on big data is determined based on user luggage data, import and export goods data, and inbound and outbound transport vehicle data.
[0007] Preferably, the data processing module includes: Data retrieval unit, used to retrieve comprehensive customs real-time data based on big data; Obtain comprehensive real-time customs data based on big data; Based on the sequential retrieval method, comprehensive real-time customs data based on big data is retrieved; Conduct consistency checks on comprehensive customs real-time data based on big data; Check whether the customs comprehensive real-time data meets the requirements based on the reasonable value range and mutual relationship of the customs comprehensive real-time data; Remove inconsistent data that is beyond the normal range, logically unreasonable or contradictory in the comprehensive real-time customs data; Process invalid and missing values in comprehensive real-time customs data based on big data; Remove invalid and missing data from customs comprehensive real-time data that are of no value to customs comprehensive information intelligent supervision; Identify customs comprehensive real-time data that is valuable for customs comprehensive information intelligent supervision; A data conversion unit is used to convert the retrieved customs comprehensive real-time data; Obtain comprehensive customs real-time data that is valuable for intelligent supervision of customs comprehensive information after retrieval; Convert valuable customs comprehensive real-time data for customs comprehensive information intelligent supervision; Determine the comprehensive real-time customs data with standardized data; A feature extraction unit, used for extracting features from the converted customs comprehensive real-time data; Obtain comprehensive real-time customs data with converted and standardized data; Extract features from the standardized customs comprehensive real-time data; Determine comprehensive customs characteristic data based on big data.
[0008] Preferably, the model training module includes: Data collection unit, used to collect comprehensive historical customs data based on big data; According to the needs of customs comprehensive information intelligent supervision, collect customs comprehensive historical data based on big data; A data division unit, used to divide the customs comprehensive historical data based on big data; Obtain comprehensive historical customs data based on big data; Divide the comprehensive historical customs data based on big data; Determine the customs intelligent supervision training set and customs intelligent supervision test set; Model training unit, used to build an AI-based customs comprehensive information intelligent supervision model; According to the needs of customs comprehensive information intelligent supervision, select a neural network framework suitable for customs comprehensive information intelligent supervision; Input the customs intelligent supervision training set into the selected neural network framework suitable for customs comprehensive information intelligent supervision; Based on the customs intelligent supervision training set, the selected neural network framework suitable for customs comprehensive information intelligent supervision is trained; Determine the customs comprehensive information intelligent supervision model based on artificial intelligence.
[0009] Preferably, the model training module further includes: A data screening module is used to collect customs comprehensive historical data based on big data according to the needs of customs comprehensive information intelligent supervision, and to process the customs comprehensive historical data to remove erroneous data and garbled data in the customs comprehensive historical data; A first data volume acquisition module, used to extract the data volume corresponding to the erroneous data in the customs comprehensive historical data; A second data volume acquisition module is used to extract the data volume of the garbled data in the customs comprehensive historical data; The data quality evaluation coefficient acquisition module is used to obtain the data quality evaluation coefficient of the customs comprehensive historical data according to the data volume corresponding to the error data and the data volume of the garbled data in the customs comprehensive historical data; wherein the data quality evaluation coefficient is obtained by the following formula: ; Among them, S represents the data quality evaluation coefficient; D w Indicates the amount of data corresponding to the erroneous data in the customs comprehensive historical data; Dm represents the amount of garbled data in the customs comprehensive historical data; D z represents the total amount of customs comprehensive historical data; n represents the number of data types contained in the error data; m represents the number of data types contained in the garbled data; D i represents the amount of erroneous data corresponding to the data type contained in the i-th type of erroneous data; T i Indicates the total amount of data types contained in the i-th type of error data; D j represents the amount of garbled data corresponding to the data type contained in the jth type of garbled data; T j represents the total amount of data types contained in the jth type of garbled data; k and r represent the nonlinear adjustment indexes corresponding to the error data and garbled data, respectively; f(Q) represents the data quality function, and the function form of the data quality function is as follows: ; Wherein, f(Q) represents the data quality function; v represents the number of data quality determination dimensions, wherein the data quality determination dimensions include data consistency, data integrity, data accuracy, data timeliness, data redundancy and data accessibility; λ i represents the weight factor corresponding to the i-th data quality judgment dimension; Q i represents the metric value corresponding to the i-th data quality judgment dimension, and the metric values corresponding to the data consistency, data integrity, data accuracy, data timeliness, data redundancy and data accessibility are consistency ratio, completion ratio, error rate, data delay reception ratio, data redundancy ratio and accessible data ratio respectively; δ i represents the correction parameter corresponding to the i-th data quality judgment dimension; t represents the preset multi-dimensional numerical comprehensive parameter threshold; A comparison module, used for comparing the data quality evaluation coefficient with a preset coefficient threshold; A data re-collection control module, used for determining that the quality of the customs comprehensive historical data is poor when the data quality evaluation coefficient is lower than a preset coefficient threshold, and controlling the data collection unit to re-collect the customs comprehensive historical data; The quality determination module is used to determine that the quality of the customs comprehensive historical data is good when the data quality evaluation coefficient is not lower than a preset coefficient threshold.
[0010] Preferably, the model training module further includes: A data quality evaluation coefficient extraction module, used for extracting the data quality evaluation coefficient after determining that the customs comprehensive historical data has good quality; A coefficient threshold extraction module, used to extract the preset coefficient threshold; A ratio range acquisition module, used to obtain the ratio range of the customs intelligent supervision training set and the customs intelligent supervision test set corresponding to the customs comprehensive historical data by using the quantitative relationship between the specific value of the data quality evaluation coefficient and the preset coefficient threshold; The upper limit and lower limit of the ratio corresponding to the ratio range are obtained by the following formula: ; Among them, U represents the upper limit of the ratio; U 0 represents the initial value of the preset upper limit of the ratio; u represents the first adjustment coefficient, and the value range of the first adjustment coefficient is 0.36-0.87; S represents the data quality evaluation coefficient; S 0 represents the preset coefficient threshold; k and r represent the nonlinear adjustment index corresponding to the error data and the garbled data respectively; ; Wherein, L represents the lower limit of the ratio; L 0 represents the initial value of the preset lower limit of the ratio; e represents the second adjustment coefficient, and the value range of the second adjustment coefficient is 0.18-0.59; S represents the data quality evaluation coefficient; S 0 represents the preset coefficient threshold; k and r represent the nonlinear adjustment indexes corresponding to the error data and garbled data respectively.
[0011] Preferably, the evaluation and optimization module includes: Test and evaluation unit, used to test and evaluate the customs comprehensive information intelligent supervision model; Obtaining an AI-based intelligent customs supervision model for comprehensive information; Based on the customs intelligent supervision test set, the customs comprehensive information intelligent supervision model was tested and evaluated; Determine the test and evaluation results of the intelligent supervision model based on customs comprehensive information; Optimization and adjustment unit, used to optimize and adjust the customs comprehensive information intelligent supervision model; Obtain test and evaluation results based on the customs comprehensive information intelligent supervision model; Conduct mining and analysis on the test and evaluation results based on the customs comprehensive information intelligent supervision model; Determine the optimal adjustment plan based on the customs comprehensive information intelligent supervision model; Optimize and adjust the customs comprehensive information intelligent supervision model based on the optimization and adjustment plan; Determine the optimal customs comprehensive information intelligent supervision model.
[0012] Preferably, the supervision analysis module includes: A data extraction unit, used to extract comprehensive customs feature data based on big data; According to the needs of intelligent supervision of customs comprehensive information, extract customs comprehensive characteristic data based on big data; Regulatory analysis unit, used to conduct regulatory analysis on customs comprehensive characteristic data; Obtain the best customs comprehensive information intelligent supervision model; Input the customs comprehensive characteristic data based on big data into the optimal customs comprehensive information intelligent supervision model; Conduct supervision and analysis on customs comprehensive characteristic data based on big data based on the optimal customs comprehensive information intelligent supervision model; Determine the results of intelligent supervision and analysis of comprehensive customs information based on artificial intelligence; Among them, the results of the intelligent supervision analysis of comprehensive customs information based on artificial intelligence include abnormal intelligent supervision of comprehensive customs information or normal intelligent supervision of comprehensive customs information; The visualization display unit is used to visualize the results of intelligent supervision and analysis of comprehensive customs information based on artificial intelligence to users.
[0013] Preferably, the intelligent management and control module includes: Strategy formulation unit, used to formulate intelligent supervision plans for customs comprehensive information; Obtain intelligent supervision and analysis results of comprehensive customs information based on artificial intelligence; Conduct mining and analysis on the results of intelligent supervision and analysis of comprehensive customs information based on artificial intelligence; Determine the intelligent supervision plan of comprehensive customs information based on artificial intelligence; Intelligent control unit, used for intelligent supervision of comprehensive customs information; Obtain an intelligent customs supervision solution based on artificial intelligence; Carry out intelligent supervision of customs comprehensive information based on the intelligent supervision solution of customs comprehensive information.
[0014] Preferably, the comprehensive customs information is intelligently supervised and the following operations are performed: Provide timely warnings for comprehensive customs information with abnormal intelligent supervision, quickly locate abnormal customs business processes, take intelligent management and control measures, and solve abnormal problems in a timely manner.
[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention collects comprehensive customs real-time data based on big data, processes the comprehensive customs real-time data, determines comprehensive customs feature data based on big data, constructs a comprehensive customs information intelligent supervision model based on artificial intelligence according to the demand for intelligent supervision of comprehensive customs information, determines the optimal comprehensive customs information intelligent supervision model by evaluating and optimizing the comprehensive customs information intelligent supervision model, performs supervision and analysis on the comprehensive customs feature data based on big data based on the optimal comprehensive customs information intelligent supervision model, determines the intelligent supervision analysis result of comprehensive customs information based on artificial intelligence, formulates an intelligent supervision plan for comprehensive customs information, and performs intelligent supervision on comprehensive customs information based on the intelligent supervision plan for comprehensive customs information, which can realize all-round monitoring and early warning of customs business, realize automation and intelligence of customs business processes, make customs supervision effect good, and improve customs supervision efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1 This is a structural diagram of the customs comprehensive information intelligent supervision system based on big data and artificial intelligence of the present invention. DETAILED DESCRIPTION
[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.
[0018] In order to solve the existing problems of not being able to conduct intelligent supervision of customs comprehensive information, not being able to achieve all-round monitoring and early warning of customs business, resulting in poor customs supervision effect and low customs supervision efficiency, please refer to Figure 1 , this embodiment provides the following technical solutions: The customs comprehensive information intelligent supervision system based on big data and artificial intelligence includes: data collection module, data processing module, model training module, evaluation and optimization module, supervision analysis module and intelligent management and control module; It should be noted that through the interactive communication between the data collection module, data processing module, model training module, evaluation and optimization module, supervision analysis module and intelligent management and control module, all-round monitoring and early warning of customs business can be achieved, and the automation and intelligentization of customs business processes can be realized, so that customs supervision is effective and customs supervision efficiency can be improved.
[0019] Among them, the data collection module is used to collect comprehensive real-time customs data based on big data; In this embodiment, the data acquisition module includes: The luggage collection unit is used to monitor the user's luggage and mail items in real time and obtain the user's luggage data; Import and export cargo collection unit, used to monitor import and export cargo in real time and obtain import and export cargo data; The inbound and outbound means collection unit is used to monitor the inbound and outbound means of transport in real time and obtain the inbound and outbound means of transport data; Among them, comprehensive real-time customs data based on big data is determined based on user luggage data, import and export goods data, and inbound and outbound transport vehicle data.
[0020] It should be noted that through real-time monitoring of user luggage and postal items, import and export goods, and inbound and outbound means of transport, it is possible to obtain user luggage data, import and export goods data, and inbound and outbound means of transport data, thereby achieving all-round monitoring of customs business and facilitating early warning in the event of anomalies.
[0021] Among them, the data processing module is used to process comprehensive real-time customs data based on big data; In this embodiment, the data processing module includes: Data retrieval unit, used to retrieve comprehensive customs real-time data based on big data; Obtain comprehensive real-time customs data based on big data; Based on the sequential retrieval method, comprehensive real-time customs data based on big data is retrieved; Conduct consistency checks on comprehensive customs real-time data based on big data; Check whether the customs comprehensive real-time data meets the requirements based on the reasonable value range and mutual relationship of the customs comprehensive real-time data; Remove inconsistent data that is beyond the normal range, logically unreasonable or contradictory in the customs comprehensive real-time data; Process invalid and missing values in comprehensive real-time customs data based on big data; Remove invalid and missing data from customs comprehensive real-time data that are of no value to customs comprehensive information intelligent supervision; Identify customs comprehensive real-time data that is valuable for customs comprehensive information intelligent supervision; A data conversion unit is used to convert the retrieved customs comprehensive real-time data; Obtain comprehensive customs real-time data that is valuable for intelligent supervision of customs comprehensive information after retrieval; Convert valuable customs comprehensive real-time data for customs comprehensive information intelligent supervision; Determine the comprehensive real-time customs data with standardized data; A feature extraction unit, used for extracting features from the converted customs comprehensive real-time data; Obtain comprehensive real-time customs data with converted and standardized data; Extract features from the standardized customs comprehensive real-time data; Determine comprehensive customs characteristic data based on big data.
[0022] Among them, the model training module is used to build an AI-based customs comprehensive information intelligent supervision model; In this embodiment, the model training module includes: Data collection unit, used to collect comprehensive historical customs data based on big data; According to the needs of customs comprehensive information intelligent supervision, collect customs comprehensive historical data based on big data; A data division unit, used to divide the customs comprehensive historical data based on big data; Obtain comprehensive historical customs data based on big data; Divide the comprehensive historical customs data based on big data; Determine the customs intelligent supervision training set and customs intelligent supervision test set; Model training unit, used to build an AI-based customs comprehensive information intelligent supervision model; According to the needs of customs comprehensive information intelligent supervision, select a neural network framework suitable for customs comprehensive information intelligent supervision; Input the customs intelligent supervision training set into the selected neural network framework suitable for customs comprehensive information intelligent supervision; Based on the customs intelligent supervision training set, the selected neural network framework suitable for customs comprehensive information intelligent supervision is trained; Determine the customs comprehensive information intelligent supervision model based on artificial intelligence.
[0023] Among them, the evaluation and optimization module is used to evaluate and optimize the customs comprehensive information intelligent supervision model; Specifically, the model training module also includes: A data screening module is used to collect customs comprehensive historical data based on big data according to the needs of customs comprehensive information intelligent supervision, and to process the customs comprehensive historical data to remove erroneous data and garbled data in the customs comprehensive historical data; A first data volume acquisition module, used to extract the data volume corresponding to the erroneous data in the customs comprehensive historical data; A second data volume acquisition module is used to extract the data volume of the garbled data in the customs comprehensive historical data; The data quality evaluation coefficient acquisition module is used to obtain the data quality evaluation coefficient of the customs comprehensive historical data according to the data volume corresponding to the error data and the data volume of the garbled data in the customs comprehensive historical data; wherein the data quality evaluation coefficient is obtained by the following formula: ; Among them, S represents the data quality evaluation coefficient; D w Indicates the amount of data corresponding to the erroneous data in the customs comprehensive historical data; D m represents the amount of garbled data in the customs comprehensive historical data; D z represents the total amount of customs comprehensive historical data; n represents the number of data types contained in the error data; m represents the number of data types contained in the garbled data; D i represents the amount of erroneous data corresponding to the data type contained in the i-th type of erroneous data; T i Indicates the total amount of data types contained in the i-th type of error data; D j represents the amount of garbled data corresponding to the data type contained in the jth type of garbled data; T j represents the total amount of data types contained in the jth type of garbled data; k and r represent the nonlinear adjustment indexes corresponding to the error data and garbled data, respectively; f(Q) represents the data quality function, and the function form of the data quality function is as follows: ; Wherein, f(Q) represents the data quality function; v represents the number of data quality determination dimensions, wherein the data quality determination dimensions include data consistency, data integrity, data accuracy, data timeliness, data redundancy and data accessibility; λ i represents the weight factor corresponding to the i-th data quality judgment dimension; Q i represents the metric value corresponding to the i-th data quality judgment dimension, and the metric values corresponding to the data consistency, data integrity, data accuracy, data timeliness, data redundancy and data accessibility are consistency ratio, completion ratio, error rate, data delay reception ratio, data redundancy ratio and accessible data ratio respectively; δ i represents the correction parameter corresponding to the i-th data quality judgment dimension; t represents the preset multi-dimensional numerical comprehensive parameter threshold; A comparison module, used for comparing the data quality evaluation coefficient with a preset coefficient threshold; A data re-collection control module, used for determining that the quality of the customs comprehensive historical data is poor when the data quality evaluation coefficient is lower than a preset coefficient threshold, and controlling the data collection unit to re-collect the customs comprehensive historical data; The quality determination module is used to determine that the quality of the customs comprehensive historical data is good when the data quality evaluation coefficient is not lower than a preset coefficient threshold.
[0024] The technical effect of the above technical solution is that through the data screening module, the system can automatically identify and remove erroneous data and garbled data in the customs comprehensive historical data, significantly improving the cleanliness and accuracy of the data set. This is crucial for subsequent data analysis and model training, because inaccurate data may lead to deviations in analysis results or degradation of model performance.
[0025] The above technical solution introduces a data quality evaluation coefficient (S), which comprehensively considers the types and quantities of erroneous data and garbled data, as well as data quality judgment dimensions (such as consistency, completeness, accuracy, etc.) through a complex calculation formula, providing a quantifiable evaluation standard for data quality. This helps to objectively evaluate the overall quality of the data set and provide strong support for subsequent data processing decisions. When the data quality evaluation coefficient is lower than the preset coefficient threshold, the system will determine that the quality of the customs comprehensive historical data is poor, and automatically control the data collection unit to re-collect data. This automated data quality control mechanism can ensure that the data input to the model always maintains a high quality level, thereby improving the efficiency and effectiveness of model training.
[0026] The formula in the above technical solution contains multiple adjustable parameters (such as nonlinear adjustment indexes k and r, weight factor λi of data quality judgment dimension, etc.), which can be adjusted and optimized according to actual needs to adapt to different application scenarios and data characteristics. This flexibility makes the technical solution have broad application prospects and strong adaptability. Through strict data screening and quality control mechanisms, the solution ensures that the customs comprehensive historical data input into the model has a high degree of accuracy and reliability. High-quality data input is one of the keys to successful model training. It can reduce noise interference and deviation accumulation during model training, thereby improving the prediction accuracy and generalization ability of the model.
[0027] In summary, the above technical solutions have significantly improved the quality of customs comprehensive historical data through automated data screening, quantitative data quality assessment and flexible data quality control mechanisms, providing a solid foundation for subsequent model training. This helps to improve the efficiency and effectiveness of model training, thereby enhancing the overall performance and reliability of the customs comprehensive information intelligent supervision system.
[0028] Specifically, the model training module also includes: A data quality evaluation coefficient extraction module, used for extracting the data quality evaluation coefficient after determining that the customs comprehensive historical data has good quality; A coefficient threshold extraction module, used to extract the preset coefficient threshold; A ratio range acquisition module, used to obtain the ratio range of the customs intelligent supervision training set and the customs intelligent supervision test set corresponding to the customs comprehensive historical data by using the quantitative relationship between the specific value of the data quality evaluation coefficient and the preset coefficient threshold; The upper limit and lower limit of the ratio corresponding to the ratio range are obtained by the following formula: ; Among them, U represents the upper limit of the ratio; U 0 represents the initial value of the preset upper limit of the ratio; u represents the first adjustment coefficient, and the value range of the first adjustment coefficient is 0.36-0.87; S represents the data quality evaluation coefficient; S 0 represents the preset coefficient threshold; k and r represent the nonlinear adjustment index corresponding to the error data and the garbled data respectively; ; Wherein, L represents the lower limit of the ratio; L 0 represents the initial value of the preset lower limit of the ratio; e represents the second adjustment coefficient, and the value range of the second adjustment coefficient is 0.18-0.59; S represents the data quality evaluation coefficient; S 0 represents the preset coefficient threshold; k and r represent the nonlinear adjustment indexes corresponding to the error data and garbled data respectively.
[0029] The technical effect of the above technical solution is: by utilizing the quantitative relationship between the data quality evaluation coefficient and the preset coefficient threshold, the solution can accurately calculate the proportion range of the customs intelligent supervision training set and the test set corresponding to the customs comprehensive historical data. This refined data set division helps to ensure the reasonable allocation of the training set and the test set during the model training process, thereby improving the efficiency and accuracy of the model training. The upper limit value and the lower limit value of the proportion in the above technical solution are dynamically calculated by a complex formula, which takes into account multiple factors such as the data quality evaluation coefficient, the preset coefficient threshold, and the nonlinear adjustment index corresponding to the erroneous data and the garbled data. This dynamic adjustment mechanism enables the proportion of the data set to be flexibly adjusted according to the actual situation of the data quality, so as to adapt to different data environments and application scenarios.
[0030] By reasonably dividing the ratio of training set and test set, the scheme helps to ensure that the model can fully learn the characteristics of the data during the training process, and can accurately evaluate the generalization ability of the model during the testing phase. This is crucial for building a customs intelligent supervision model with good generalization ability. Since the scheme takes into account the quantitative relationship between the data quality evaluation coefficient and the preset coefficient threshold, it can avoid the problem of unstable model training caused by poor data quality when dividing the data set. This helps to ensure the smooth progress of model training and improve the stability and reliability of the model. Through refined data set division and dynamic adjustment mechanism, the scheme can maximize the use of effective information in customs comprehensive historical data. This helps to improve the utilization of data and reduce the waste of information caused by improper data division.
[0031] In summary, the above technical solutions provide strong support for the training of customs intelligent supervision models through refined data set division, dynamic adjustment of data set ratio, improvement of model generalization ability, enhancement of model training stability and improvement of data utilization. This helps to build a more accurate, stable and reliable customs intelligent supervision system, thereby improving the efficiency and accuracy of customs supervision.
[0032] In this embodiment, the evaluation and optimization module includes: Test and evaluation unit, used to test and evaluate the customs comprehensive information intelligent supervision model; Obtaining an AI-based intelligent customs supervision model for comprehensive information; Based on the customs intelligent supervision test set, the customs comprehensive information intelligent supervision model was tested and evaluated; Determine the test and evaluation results of the intelligent supervision model based on customs comprehensive information; Optimization and adjustment unit, used to optimize and adjust the customs comprehensive information intelligent supervision model; Obtain test and evaluation results based on the customs comprehensive information intelligent supervision model; Conduct mining and analysis on the test and evaluation results based on the customs comprehensive information intelligent supervision model; Determine the optimal adjustment plan based on the customs comprehensive information intelligent supervision model; Optimize and adjust the customs comprehensive information intelligent supervision model based on the optimization and adjustment plan; Determine the optimal customs comprehensive information intelligent supervision model.
[0033] Among them, the supervision analysis module is used to conduct supervision analysis on customs comprehensive characteristic data and determine the intelligent supervision analysis results of customs comprehensive information based on artificial intelligence; In this embodiment, the supervision analysis module includes: A data extraction unit, used to extract comprehensive customs feature data based on big data; According to the needs of intelligent supervision of customs comprehensive information, extract customs comprehensive characteristic data based on big data; Regulatory analysis unit, used to conduct regulatory analysis on customs comprehensive characteristic data; Obtain the best customs comprehensive information intelligent supervision model; Input the customs comprehensive characteristic data based on big data into the optimal customs comprehensive information intelligent supervision model; Conduct supervision and analysis on customs comprehensive characteristic data based on big data based on the optimal customs comprehensive information intelligent supervision model; Determine the results of intelligent supervision and analysis of comprehensive customs information based on artificial intelligence; Among them, the results of the intelligent supervision analysis of comprehensive customs information based on artificial intelligence include abnormal intelligent supervision of comprehensive customs information or normal intelligent supervision of comprehensive customs information; The visualization display unit is used to visualize the results of intelligent supervision and analysis of comprehensive customs information based on artificial intelligence to users.
[0034] Among them, the intelligent management and control module is used to conduct intelligent supervision of comprehensive customs information.
[0035] In this embodiment, the intelligent management and control module includes: Strategy formulation unit, used to formulate intelligent supervision plans for customs comprehensive information; Obtain intelligent supervision and analysis results of comprehensive customs information based on artificial intelligence; Conduct mining and analysis on the results of intelligent supervision and analysis of comprehensive customs information based on artificial intelligence; Determine the intelligent supervision plan of comprehensive customs information based on artificial intelligence; Intelligent control unit, used for intelligent supervision of comprehensive customs information; Obtain an intelligent customs supervision solution based on artificial intelligence; Carry out intelligent supervision of customs comprehensive information based on the intelligent supervision solution of customs comprehensive information.
[0036] In this embodiment, the comprehensive customs information is supervised intelligently, and the following operations are performed: Provide timely warnings for comprehensive customs information with abnormal intelligent supervision, quickly locate abnormal customs business processes, take intelligent management and control measures, and solve abnormal problems in a timely manner.
[0037] Therefore, by collecting comprehensive customs real-time data based on big data, processing the comprehensive customs real-time data, determining the comprehensive customs feature data based on big data, and according to the needs of customs comprehensive information intelligent supervision, constructing a customs comprehensive information intelligent supervision model based on artificial intelligence, and evaluating and optimizing the customs comprehensive information intelligent supervision model, determining the optimal customs comprehensive information intelligent supervision model, and conducting supervision and analysis on the customs comprehensive feature data based on big data based on the optimal customs comprehensive information intelligent supervision model, determining the results of customs comprehensive information intelligent supervision analysis based on artificial intelligence, and formulating a customs comprehensive information intelligent supervision plan, and conducting intelligent supervision on customs comprehensive information based on the customs comprehensive information intelligent supervision plan, it is possible to achieve all-round monitoring and early warning of customs business, realize the automation and intelligence of customs business processes, make customs supervision more effective, and improve customs supervision efficiency.
[0038] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0039] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. Customs comprehensive information intelligent supervision system based on big data and artificial intelligence, characterized by: include: Data collection module, used to collect comprehensive real-time customs data based on big data; Data processing module, used to process comprehensive real-time customs data based on big data; Model training module, used to build an AI-based customs comprehensive information intelligent supervision model; Evaluation and optimization module, used to evaluate and optimize the customs comprehensive information intelligent supervision model; The supervision and analysis module is used to conduct supervision and analysis on the comprehensive characteristic data of customs and determine the intelligent supervision and analysis results of comprehensive customs information based on artificial intelligence; Intelligent management and control module is used to conduct intelligent supervision of comprehensive customs information.
2. The customs comprehensive information intelligent supervision system based on big data and artificial intelligence according to claim 1 is characterized in that: The data acquisition module comprises: The luggage collection unit is used to monitor the user's luggage and mail items in real time and obtain the user's luggage data; Import and export cargo collection unit, used to monitor import and export cargo in real time and obtain import and export cargo data; The inbound and outbound means collection unit is used to monitor the inbound and outbound means of transport in real time and obtain the inbound and outbound means of transport data; Among them, comprehensive real-time customs data based on big data is determined based on user luggage data, import and export goods data, and inbound and outbound transport vehicle data.
3. The customs comprehensive information intelligent supervision system based on big data and artificial intelligence according to claim 2 is characterized in that: The data processing module comprises: Data retrieval unit, used to retrieve comprehensive customs real-time data based on big data; Obtain comprehensive real-time customs data based on big data; Based on the sequential retrieval method, comprehensive real-time customs data based on big data is retrieved; Conduct consistency checks on comprehensive real-time customs data based on big data; Check whether the customs comprehensive real-time data meets the requirements based on the reasonable value range and mutual relationship of the customs comprehensive real-time data; Remove inconsistent data that is beyond the normal range, logically unreasonable or contradictory in the customs comprehensive real-time data; Process invalid and missing values in comprehensive real-time customs data based on big data; Remove invalid and missing data from customs comprehensive real-time data that are of no value to customs comprehensive information intelligent supervision; Identify customs comprehensive real-time data that is valuable for customs comprehensive information intelligent supervision; A data conversion unit, used to convert the retrieved customs comprehensive real-time data; Obtain comprehensive real-time customs data that is valuable for intelligent supervision of comprehensive customs information after retrieval; Convert valuable customs comprehensive real-time data for customs comprehensive information intelligent supervision; Determine the comprehensive real-time customs data with standardized data; A feature extraction unit, used for extracting features from the converted customs comprehensive real-time data; Obtain comprehensive real-time customs data with converted and standardized data; Extract features from the standardized customs comprehensive real-time data; Determine comprehensive customs characteristic data based on big data.
4. The customs comprehensive information intelligent supervision system based on big data and artificial intelligence according to claim 3 is characterized in that: The model training module includes: Data collection unit, used to collect comprehensive historical customs data based on big data; According to the needs of customs comprehensive information intelligent supervision, collect customs comprehensive historical data based on big data; A data division unit, used to divide the customs comprehensive historical data based on big data; Obtain comprehensive historical customs data based on big data; Divide the comprehensive historical customs data based on big data; Determine the customs intelligent supervision training set and customs intelligent supervision test set; Model training unit, used to build an AI-based customs comprehensive information intelligent supervision model; According to the needs of customs comprehensive information intelligent supervision, select a neural network framework suitable for customs comprehensive information intelligent supervision; Input the customs intelligent supervision training set into the selected neural network framework suitable for customs comprehensive information intelligent supervision; Based on the customs intelligent supervision training set, the selected neural network framework suitable for customs comprehensive information intelligent supervision is trained; Determine the customs comprehensive information intelligent supervision model based on artificial intelligence.
5. The customs comprehensive information intelligent supervision system based on big data and artificial intelligence according to claim 4 is characterized in that: The model training module further includes: A data screening module is used to collect customs comprehensive historical data based on big data according to the needs of customs comprehensive information intelligent supervision, and to process the customs comprehensive historical data to remove erroneous data and garbled data in the customs comprehensive historical data; A first data volume acquisition module, used to extract the data volume corresponding to the erroneous data in the customs comprehensive historical data; A second data volume acquisition module is used to extract the data volume of the garbled data in the customs comprehensive historical data; The data quality evaluation coefficient acquisition module is used to obtain the data quality evaluation coefficient of the customs comprehensive historical data according to the data volume corresponding to the error data and the data volume of the garbled data in the customs comprehensive historical data; wherein the data quality evaluation coefficient is obtained by the following formula: ; Among them, S represents the data quality evaluation coefficient; D w Indicates the amount of data corresponding to the erroneous data in the customs comprehensive historical data; D m represents the amount of garbled data in the customs comprehensive historical data; D z represents the total amount of customs comprehensive historical data; n represents the number of data types contained in the error data; m represents the number of data types contained in the garbled data; D i represents the amount of erroneous data corresponding to the data type contained in the i-th type of erroneous data; T i Indicates the total amount of data types contained in the i-th type of error data; D j represents the amount of garbled data corresponding to the data type contained in the jth type of garbled data; T j represents the total amount of data types contained in the jth type of garbled data; k and r represent the nonlinear adjustment indexes corresponding to the error data and garbled data, respectively; f(Q) represents the data quality function, and the function form of the data quality function is as follows: ; Wherein, f(Q) represents the data quality function; v represents the number of data quality determination dimensions, wherein the data quality determination dimensions include data consistency, data integrity, data accuracy, data timeliness, data redundancy and data accessibility; λ i represents the weight factor corresponding to the i-th data quality judgment dimension; Q i represents the metric value corresponding to the i-th data quality judgment dimension, and the metric values corresponding to the data consistency, data integrity, data accuracy, data timeliness, data redundancy and data accessibility are consistency ratio, completion ratio, error rate, data delay reception ratio, data redundancy ratio and accessible data ratio respectively; δ i represents the correction parameter corresponding to the i-th data quality judgment dimension; t represents the preset multi-dimensional numerical comprehensive parameter threshold; A comparison module, used for comparing the data quality evaluation coefficient with a preset coefficient threshold; A data re-collection control module, used for determining that the quality of the customs comprehensive historical data is poor when the data quality evaluation coefficient is lower than a preset coefficient threshold, and controlling the data collection unit to re-collect the customs comprehensive historical data; The quality determination module is used to determine that the quality of the customs comprehensive historical data is good when the data quality evaluation coefficient is not lower than a preset coefficient threshold.
6. The customs comprehensive information intelligent supervision system based on big data and artificial intelligence according to claim 5 is characterized in that: The model training module further includes: A data quality evaluation coefficient extraction module, used for extracting the data quality evaluation coefficient after determining that the customs comprehensive historical data has good quality; A coefficient threshold extraction module, used to extract the preset coefficient threshold; A ratio range acquisition module, used to obtain the ratio range of the customs intelligent supervision training set and the customs intelligent supervision test set corresponding to the customs comprehensive historical data by using the quantitative relationship between the specific value of the data quality evaluation coefficient and the preset coefficient threshold; The upper limit and lower limit of the ratio corresponding to the ratio range are obtained by the following formula: ; Among them, U represents the upper limit of the ratio; U0 represents the initial value of the preset upper limit of the ratio; u represents the first adjustment coefficient, and the value range of the first adjustment coefficient is 0.36-0.87; S represents the data quality evaluation coefficient; S0 represents the preset coefficient threshold; k and r represent the nonlinear adjustment indexes corresponding to the error data and the garbled data respectively; ; Among them, L represents the lower limit of the ratio; L0 represents the initial value of the preset lower limit of the ratio; e represents the second adjustment coefficient, and the value range of the second adjustment coefficient is 0.18-0.59; S represents the data quality evaluation coefficient; S0 represents the preset coefficient threshold; k and r represent the nonlinear adjustment indexes corresponding to the erroneous data and garbled data respectively.
7. The customs comprehensive information intelligent supervision system based on big data and artificial intelligence according to claim 6 is characterized in that: The evaluation and optimization module includes: Test and evaluation unit, used to test and evaluate the customs comprehensive information intelligent supervision model; Obtaining an AI-based intelligent customs supervision model for comprehensive information; Based on the customs intelligent supervision test set, the customs comprehensive information intelligent supervision model was tested and evaluated; Determine the test and evaluation results of the intelligent supervision model based on customs comprehensive information; Optimization and adjustment unit, used to optimize and adjust the customs comprehensive information intelligent supervision model; Obtain test and evaluation results based on the customs comprehensive information intelligent supervision model; Conduct mining and analysis on the test and evaluation results based on the customs comprehensive information intelligent supervision model; Determine the optimal adjustment plan based on the customs comprehensive information intelligent supervision model; Optimize and adjust the customs comprehensive information intelligent supervision model based on the optimization and adjustment plan; Determine the optimal customs comprehensive information intelligent supervision model.
8. The customs comprehensive information intelligent supervision system based on big data and artificial intelligence according to claim 7 is characterized in that: The supervision analysis module includes: A data extraction unit, used to extract comprehensive customs feature data based on big data; According to the needs of intelligent supervision of customs comprehensive information, extract customs comprehensive characteristic data based on big data; Regulatory analysis unit, used to conduct regulatory analysis on customs comprehensive characteristic data; Obtain the best customs comprehensive information intelligent supervision model; Input the customs comprehensive characteristic data based on big data into the optimal customs comprehensive information intelligent supervision model; Conduct supervision and analysis on customs comprehensive characteristic data based on big data based on the optimal customs comprehensive information intelligent supervision model; Determine the results of intelligent supervision and analysis of comprehensive customs information based on artificial intelligence; Among them, the results of the intelligent supervision analysis of comprehensive customs information based on artificial intelligence include abnormal intelligent supervision of comprehensive customs information or normal intelligent supervision of comprehensive customs information; The visualization display unit is used to visualize the results of intelligent supervision and analysis of comprehensive customs information based on artificial intelligence to users.
9. The customs comprehensive information intelligent supervision system based on big data and artificial intelligence according to claim 8 is characterized in that: The intelligent management and control module includes: Strategy formulation unit, used to formulate intelligent supervision plans for customs comprehensive information; Obtain intelligent supervision and analysis results of comprehensive customs information based on artificial intelligence; Conduct mining and analysis on the results of intelligent supervision and analysis of comprehensive customs information based on artificial intelligence; Determine the intelligent supervision plan of comprehensive customs information based on artificial intelligence; Intelligent control unit, used for intelligent supervision of comprehensive customs information; Obtain an intelligent customs supervision solution based on artificial intelligence; Carry out intelligent supervision of customs comprehensive information based on the intelligent supervision solution of customs comprehensive information.
10. The customs comprehensive information intelligent supervision system based on big data and artificial intelligence according to claim 9 is characterized in that: Carry out intelligent supervision of comprehensive customs information and perform the following operations: Provide timely warnings for comprehensive customs information with abnormal intelligent supervision, quickly locate abnormal customs business processes, take intelligent management and control measures, and solve abnormal problems in a timely manner.
Citation Information
Patent Citations
Data quality assessment method and system
CN114926008A
Analysis processing method for customs clearance data monitoring
CN117112647A
Comprehensive information intelligent supervision system and method applying big data technology
CN117455247A
Rotation machine equipment fault diagnosis system and method based on multi-dimensional diagnosis
CN118605474A
Cited By
Water area ecological environment supervision system based on artificial intelligence
CN119363796A
Block chain data storage system of customs intelligent seed cabinet
CN120030606A
Big data processing analysis system based on artificial intelligence
CN120316001A
