Resource integration method and device based on customs data, equipment and storage medium

Abstract

The application relates to a customs data-based resource integration method, device and equipment and a storage medium, which comprises the following steps: obtaining historical export data corresponding to each current agent subject based on customs data, and cleaning the historical export data to obtain first export analysis data; extracting data features based on the first export analysis data; constructing a multi-dimensional agent capability portrait based on the data features, and calculating a comprehensive competitiveness index corresponding to each current agent subject based on the multi-dimensional agent capability portrait; outputting a resource allocation scheme based on the comprehensive competitiveness index; obtaining target market import demand data; establishing an identification area based on the multi-dimensional agent capability portrait and the target market import demand data; generating a supply-demand gap heat map based on the identification area; obtaining candidate agent subject information; and matching the supply-demand gap heat map and the candidate agent subject information to generate an expansion recommendation list. The application has the effect of efficiently processing data in a systematic manner.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus, device and storage medium for resource integration based on customs data. Background Technology

[0002] With the increasing frequency of global trade activities, customs import and export data has become a core information resource reflecting regional economic vitality, enterprise export capabilities, and international market supply and demand dynamics. Currently, foreign trade enterprises, supply chain management platforms, and government service agencies are generally trying to use customs data to assess the business capabilities of their agents and to conduct market expansion and resource allocation accordingly.

[0003] However, customs import and export data is large in volume, and the actual situations in different markets and regions are complex. Existing technologies cannot systematically integrate, evaluate, and allocate resources based on the existing complex data. Summary of the Invention

[0004] In order to systematically and efficiently process data, this application provides a resource integration method, apparatus, equipment and storage medium based on customs data.

[0005] Firstly, the resource integration method based on customs data provided in this application adopts the following technical solution: A resource integration method based on customs data includes: Based on customs data, historical export data corresponding to each current agent entity is obtained and the historical export data is cleaned to obtain the first export analysis data. Data features are extracted based on the first export analysis data; The data features include at least the distribution of export destination countries, the proportion of export volume, and the main product categories; Based on the data features, a multi-dimensional agent capability profile is constructed, and based on the multi-dimensional agent capability profile, the comprehensive competitiveness index corresponding to each current agent entity is calculated. Resource allocation schemes are output based on the aforementioned comprehensive competitiveness index; Obtain import demand data from the target market; An identification region is established based on the multi-dimensional agent capability profile and the target market import demand data; The identified areas include areas with idle resources, market gaps, and high-potential areas. A supply-demand gap heatmap is generated based on the identified region; Obtain candidate agent entity information; An expanded recommendation list is generated by matching the supply and demand gap heat map with the candidate agent entity information.

[0006] By adopting the above technical solution, the electronic device obtains historical export data for each current agent based on customs data. This historical export data is cleaned to remove noise and redundant information, resulting in standardized first-stage export analysis data. This provides an accurate and reliable data foundation for subsequent data processing, improving data quality and usability. Subsequently, data features are extracted based on this first-stage export analysis data to construct a multi-dimensional agent capability profile. The comprehensive competitiveness index corresponding to the current agent is calculated, converting comprehensive capabilities into comparable quantitative values, reducing subjective judgment bias, and achieving accurate quantitative assessment of the agent. Then, a resource allocation plan is output based on the comprehensive competitiveness index, resulting in a more reasonable resource allocation plan that improves the overall utilization efficiency of export resources and reduces trial-and-error costs. Combining target market import demand data with the agent capability profile, identification areas are defined and a supply-demand gap heatmap is generated. This provides a visual representation of the supply and demand market situation, offering a more intuitive basis and improving the efficiency of understanding market conditions. Based on the automatic matching of the supply-demand gap heatmap and candidate agent information, an expansion recommendation list is generated. This helps users quickly locate suitable agents for entering new markets or filling gaps, shortening the response cycle, improving efficiency, and increasing the success rate of expansion. The system achieves automated and intelligent resource integration and decision support for customs data, significantly improving operational efficiency and market adaptability.

[0007] Optionally, a multi-dimensional agent capability profile is constructed based on the data features, and a comprehensive competitiveness index is calculated for each current agent entity based on the multi-dimensional agent capability profile, including: Calculate the regional coverage entropy value, market share dynamic value, and the compatibility score between the current agent and the product category; Calculate the average fit score based on the fit score; The regional coverage entropy value, the dynamic value of market share, and the average fit score are normalized. The comprehensive competitiveness index is calculated based on the normalized regional coverage entropy, the dynamic value of market share, and the average fit score.

[0008] By employing the aforementioned technical solutions, this study calculates indicators across different dimensions, including regional coverage entropy, market share dynamics, and fit score. This comprehensively reflects the performance of the agent in terms of market breadth, market share trends, and product expertise, avoiding evaluation distortions caused by relying on a single dimension. A higher entropy value indicates a wider and more balanced distribution of the agent's export destinations, lower dependence on a single market, and stronger resilience to regional market fluctuations. Market share dynamics reflect the agent's share changes in the target market, reflecting not only its current position but also the time period in which the agent operates. The fit score quantifies the agent's expertise in product categories, preventing erroneous recommendations from highly competitive agents with mismatched products, thus improving the effectiveness of subsequent matching. The final calculated comprehensive competitiveness index directly reflects the agent's competitiveness and quantifies the indicators, facilitating automatic processing by electronic devices for agent classification, allocation scheme formulation, and expanded recommendation list generation, significantly improving decision-making efficiency and objectivity.

[0009] Optionally, establishing the identification region based on the multi-dimensional agent capability profile and the target market import demand data includes: Based on the multi-dimensional agent capability profile, the export supply capability of each current agent entity to each target market is extracted; Obtain historical import demand data and macroeconomic indicators for the target market; Obtain a time series forecasting model; The historical import demand data and the macroeconomic indicators are input into the time series forecasting model to estimate the projected product demand in the target market within a future time window. The supply-demand deviation value is calculated based on the export supply capacity and the projected demand for the product. Obtain the deviation threshold; The deviation category of the target market is determined based on the supply-demand deviation value and the deviation threshold. The identification region is established based on the deviation category.

[0010] By employing the aforementioned technical solution, the export supply capacity of the agency entity to the target market is extracted from the agency capability profile and directly compared with the product forecast demand estimated based on a time series forecasting model. This transforms the abstract supply and demand relationship into a calculable numerical deviation, providing a quantitative basis for subsequent analysis. In the forecasting process, not only historical import demand data but also macroeconomic indicators are incorporated, enabling the time series forecasting model to capture the impact of economic cycles and market environment changes on import demand, significantly improving the reliability and foresight of demand forecasts. The calculated supply-demand deviation value reflects the difference between supply and demand, allowing users to intuitively understand the supply-demand relationship. Furthermore, based on the supply-demand deviation value, target markets can be classified and identified, facilitating targeted processing and the development of different strategies for different types of markets.

[0011] Optionally, determining the deviation category of the target market based on the supply-demand deviation value and the deviation threshold includes: Obtain the overall supply and demand volatility of the industry and the historical matching success rate; The deviation threshold includes multiple dynamic thresholds; The dynamic threshold is periodically adjusted based on the overall supply and demand volatility of the industry and the historical matching success rate. After establishing the recognition region based on the deviation category, the method further includes: Extract the idle resource area; Based on the resource idle area, obtain the corresponding current agent and use the obtained current agent as the first agent; Based on the first agent, the corresponding comprehensive competitiveness index is obtained and the comprehensive competitiveness index is used as the first comprehensive competitiveness index; Sort the first comprehensive competitiveness index in descending order and generate a first candidate agent list; The ratio is calculated based on the projected demand for the product in the market gap area and the export supply capacity. A priority list for filling gaps is generated based on the ratio; The first candidate agent list and the gap-filling priority list are associated to generate a resource pre-allocation scheme.

[0012] By adopting the above technical solution, two feedback indicators—the overall industry supply and demand volatility and the historical matching success rate—are introduced to periodically adjust the deviation threshold. The dynamic threshold is more adaptable to changes, reducing the possibility of misclassification due to market fluctuations and enabling automated adjustments, thus reducing lag and improving efficiency. After establishing the identification area, resource-idle areas are extracted. These areas indicate that the export supply capacity of existing agents significantly exceeds market demand, resulting in resource waste. Therefore, agents are ranked based on their competitiveness. Then, the ratio of the predicted product demand in market gap areas to the export capacity of each agent is calculated, generating a gap-filling priority list. These two lists are then linked to achieve resource complementarity and generate a resource pre-allocation plan. This automatically guides agents with excess supply capacity to areas with gaps to fill them, addressing both the offsetting problem in resource-constrained areas and filling demand in market gap areas, achieving dynamic supply-side balance.

[0013] Optionally, the step of matching the supply-demand gap heatmap and the candidate agent information to generate an expanded recommendation list includes: A candidate agent pool is constructed based on the candidate agent entity information; Based on the candidate agent entity information, extract the corresponding business characteristics; Obtain the collaborative filtering model; The operational characteristics are input into the collaborative filtering model to calculate the similarity between the candidate agent and the market gap area; The candidate agent entities are ranked based on the similarity and extended recommendations are generated. Based on the extended recommendation, generate recommendation reason text; The recommendation reason text includes the association information between the candidate agent and the blank area.

[0014] By adopting the above technical solution, electronic devices construct a unified candidate agent pool based on agent entity information, achieving data integration. Then, the operational characteristics of the candidate agents are extracted, and the similarity between the candidate agents and market gaps is calculated using these characteristics and a collaborative filtering model. Through similarity calculation, implicit connections between the two can be uncovered, improving the coverage and fit of recommendations. Furthermore, the quantified similarity can be used for ranking, generating expanded recommendations based on the ranking results, along with corresponding recommendation reason text. This allows users to not only quickly obtain more suitable agents but also understand the reasons for their suitability.

[0015] Optionally, the method further includes: Data within a first preset period is extracted based on the historical export data, and the extracted data is used as the first calculation data; A matching quality score is obtained by calculating based on the first calculation data; Determine whether the matching quality score is less than a preset matching threshold; If the matching quality score is less than the preset matching threshold, the current matching data is recorded and feedback adjustment data is output.

[0016] By employing the above technical solution, the first calculation data is extracted, and then a matching quality score is calculated. This transforms the abstract concept of matching quality into a comparable quantitative indicator, allowing users to more intuitively understand the matching situation. The system automatically judges based on the matching quality score and outputs data according to the judgment result. Specifically, when the matching quality is low, the matching result is judged as poor, so feedback adjustment data is output, while the current matching data is recorded. This matching process leaves a trace so that users can understand past experiences when matching occurs again, and it also outputs feedback adjustment data so that users can adjust their matching situation to improve matching quality.

[0017] Optionally, the method further includes: A sliding time window of a preset length is dynamically maintained based on the current time; Collect customs export data within the sliding time window; Calculate the distribution change between the current generation of the multidimensional agent capability profile and the previous generation of the multidimensional agent capability profile corresponding to the current agent subject at a first preset time interval; Calculate the data drift based on the aforementioned distribution change. Determine whether the data drift amount is greater than a preset drift threshold; When the data drift amount is greater than the preset drift threshold, an adjustment parameter is output based on the data drift amount.

[0018] By adopting the above technical solution, the electronic device collects customs export data within a sliding time window, focusing on customs export data in the recent period. Then, it calculates the distribution change between the current multidimensional agent capability profile and the previous generation multidimensional agent profile at a first preset time interval. Based on the distribution change, it calculates the data drift, combines the changes of multidimensional features into a single quantitative value, and makes an autonomous judgment. When the data drift exceeds a preset drift threshold, it indicates that the multidimensional agent profile is inaccurate, which will lead to inaccurate subsequent processing. Therefore, the electronic device outputs adjustment parameters based on the data drift, thereby adaptively adjusting the output parameters based on the degree of drift, reducing data drift and improving the accuracy of the results.

[0019] Secondly, the end-to-end wastewater treatment prediction and control device provided in this application adopts the following technical solution: Thirdly, the electronic device provided in this application adopts the following technical solution: An electronic device includes a processor coupled to a memory; the processor is configured to execute a computer program stored in the memory such that the electronic device performs the method as described in the first aspect.

[0020] Fourthly, the computer-readable storage medium provided in this application adopts the following technical solution: A computer-readable storage medium includes a computer program or instructions that, when executed on a computer, cause the computer to perform the method as described in the first aspect.

[0021] In summary, this application includes at least one of the following beneficial technical effects: The electronic device acquires historical export data for each current agent based on customs data. This historical export data is cleaned to remove noise and redundant information, resulting in standardized first-stage export analysis data. This provides an accurate and reliable data foundation for subsequent data processing, improving data quality and usability. Subsequently, data features are extracted from this first-stage export analysis data to construct a multi-dimensional agent capability profile. The comprehensive competitiveness index corresponding to the current agent is calculated, converting comprehensive capabilities into comparable quantitative values, reducing subjective judgment bias, and achieving accurate quantitative assessment of the agent. Based on the comprehensive competitiveness index, a resource allocation plan is output, resulting in a more rational resource allocation plan that improves the overall utilization efficiency of export resources and reduces trial-and-error costs. Combining target market import demand data with the agent capability profile, identification areas are defined, and a supply-demand gap heatmap is generated. This provides a visual representation of the supply and demand market situation, offering a more efficient way to understand market conditions. Based on the automatic matching of the supply-demand gap heatmap and candidate agent information, an expansion recommendation list is generated. This helps users quickly locate suitable agents for entering new markets or filling gaps, shortening the response cycle, improving efficiency, and increasing the success rate of expansion. The system achieves automated and intelligent resource integration and decision support for customs data, significantly improving operational efficiency and market adaptability. This system calculates indicators across different dimensions, including regional coverage entropy, market share dynamics, and fit score, to comprehensively reflect the performance of agents in terms of market breadth, market share trends, and product expertise, avoiding evaluation distortions caused by relying on a single dimension. A higher entropy value indicates a wider and more balanced distribution of export destinations for the corresponding agent, lower dependence on a single market, and stronger resilience to regional market fluctuations. Market share dynamics reflect the agent's share changes in the target market, reflecting not only its current position but also the time period in which the agent operates. The fit score quantifies the agent's expertise in the product category, avoiding erroneous recommendations from highly competitive agents with mismatched products, thus improving the effectiveness of subsequent matching. Finally, a comprehensive competitiveness index is calculated, which intuitively reflects the agent's competitiveness and quantifies the indicators. This facilitates automatic processing by electronic devices for agent classification, allocation scheme formulation, and expanded recommendation list generation, significantly improving decision-making efficiency and objectivity. Attached Figure Description

[0022] Figure 1 This is a flowchart of a resource integration method based on customs data, as described in this application.

[0023] Figure 2 This is a block diagram of a resource integration device based on customs data according to an embodiment of this application.

[0024] Figure 3 This is a block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0025] This specific embodiment is merely an explanation of this application and is not intended to limit it. Users skilled in the art can make modifications to this embodiment without contributing any inventive step after reading this specification, but such modifications are protected by patent law as long as they fall within the scope of the claims of this application.

[0026] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by users of ordinary skills in the art without creative effort are within the scope of protection of this application.

[0027] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0028] This application discloses a resource integration method based on customs data. This resource integration method based on customs data can be executed by an electronic device. The electronic device can be a server or a terminal device. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smartphone, tablet computer, desktop computer, etc., but is not limited to these.

[0029] This application discloses a resource integration method based on customs data. (Refer to...) Figure 1 The main processes of a resource integration method based on customs data are described below (S100~S900): Step S100: Based on customs data, obtain historical export data corresponding to each current agent entity and clean the historical export data to obtain the first export analysis data; Step S200: Extract data features based on the first exit analysis data; Data characteristics should include at least the distribution of export destinations, the proportion of export volume, and the categories of main products. Step S300: Construct a multi-dimensional agent capability profile based on data features and calculate the comprehensive competitiveness index corresponding to each current agent entity based on data features; Step S400: Output a resource allocation plan based on the comprehensive competitiveness index; Step S500: Obtain import demand data from the target market; Step S600: Establish an identification region based on the multi-dimensional agent capability profile and target market import demand data; The identified areas include areas with idle resources, areas with market gaps, and areas with high potential. Step S700: Generate a heat map of supply and demand gaps based on the identified regions; Step S800: Obtain candidate agent entity information; Step S900: Match the supply and demand gap heatmap and candidate agent information to generate an expanded recommendation list.

[0030] The electronic device acquires historical export data for each current agent entity based on customs data. This historical export data is then cleaned to obtain primary export analysis data. The cleaning process includes outlier removal, duplicate record deduplication, and data standardization to obtain standardized primary export analysis data. Data features are extracted from this primary export analysis data, and a multi-dimensional agent capability profile is constructed based on these features. The electronic device then uses this multi-dimensional agent capability profile to calculate the comprehensive competitiveness index for each current agent entity. Specifically, the electronic device calculates the comprehensive competitiveness index based on at least three dimensions of features. Finally, the electronic device can output a resource allocation plan based on the comprehensive competitiveness index.

[0031] Simultaneously, the electronic device acquires import demand data from the target market. Then, based on a multi-dimensional profile of agency capabilities and the target market import demand data, it establishes identification regions. These regions include areas with idle resources, market gaps, and high-potential areas. The electronic device generates a supply-demand gap heatmap based on these regions, effectively dividing the target market and generating a heatmap to more easily understand its type and situation. Next, the electronic device acquires information on candidate agents and matches this information with the supply-demand gap heatmap to generate an expanded recommendation list. In other words, the electronic device can perform a certain degree of matching and recommendation based on different target market conditions and candidate agent information to better select suitable agents.

[0032] Specifically, the comprehensive competitiveness index for each current agent is calculated based on data characteristics, including: calculating the regional coverage entropy, market share dynamics, and the fit score between the current agent and the product category; calculating the average fit score based on the fit score; normalizing the regional coverage entropy, market share dynamics, and average fit score; and calculating the comprehensive competitiveness index based on the normalized regional coverage entropy, market share dynamics, and average fit score.

[0033] The specific calculation formula is as follows: The formula for calculating the regional cover entropy is: ; in, This represents the region's coverage entropy value. For the current agent In the market The proportion of its exports to its total exports; Total number of markets; Here, the regional coverage entropy value is used to characterize the market layout equilibrium of the current agent entity; The formula for calculating the dynamic value of market share is: ; in, This represents a dynamic market share value. For the current agent The proportion of its export value to the industry's total export value; This represents the average annual growth rate of exports. The fit score is: ; in, Assign a score based on fit; For the current agent Export products The value of the goods; For the current agent The total value of all exported products; The fit score can be further calculated, and the results obtained from the above calculations can be used to... Substitute into the following formula: ; in, The score is the intermediate fit. For the current agent Export products The growth rate over the most recent year; For the current agent Credit rating; Then the calculated intermediate fit score is assigned to .

[0034] When further calculating the fit score... The calculated value is used as an intermediate value.

[0035] The formula for calculating the average fitness score is: ; The formula for calculating the comprehensive competitiveness index is: ; in, This is the comprehensive competitiveness index; The region cover entropy value after normalization, i.e., the value after normalization. use express; This represents the dynamic market share value after normalization. The average fitness score after normalization; Specifically, the identification region is established based on a multi-dimensional agent capability profile and target market import demand data. This includes: extracting the export supply capacity of each current agent to each target market based on the multi-dimensional agent capability profile; obtaining historical import demand data and macroeconomic indicators for the target market; obtaining a time series forecasting model; inputting historical import demand data and macroeconomic indicators into the time series forecasting model to estimate the projected product demand in the target market within a future time window; calculating the supply-demand deviation value based on export supply capacity and projected product demand; obtaining a deviation threshold; determining the deviation category of the target market based on the supply-demand deviation value and the deviation threshold; and establishing the identification region based on the deviation category.

[0036] The time series prediction model can be either a differentially integrated moving average autoregressive model or a long short-term memory network model. These two models have different focuses and can be selected for different scenarios. This embodiment does not modify the differentially integrated moving average autoregressive model or the long short-term memory network model; only adaptive matching is required.

[0037] The export supply capacity of each current agent to each target market is calculated using the historical average export value and growth rate of the current agent in the target market. The specific calculation formula is as follows: ; in, For the current agent In the target market Export supply capacity; For the current agent In the target market Historical average export value; Current agent The average export growth rate; The process deviation value can be calculated by subtracting the export supply capacity and the predicted product demand. The deviation threshold includes multiple dynamic thresholds; for example, the dynamic thresholds are as follows: The supply-demand deviation is greater than hour, If the value is greater than zero, it is considered a resource-idle area because the export supply capacity far exceeds the demand; when the absolute value of the supply-demand deviation is less than zero... hour, A value greater than zero indicates that the difference between export supply capacity and demand is small, thus classifying it as a supply-demand equilibrium region; similarly, when the supply-demand deviation is less than zero... When this occurs, it is determined to be a market gap area. Less than zero.

[0038] Specifically, the deviation category of the target market is determined based on the supply and demand deviation value and deviation threshold, including: the deviation threshold includes multiple dynamic thresholds; the overall supply and demand volatility of the industry and the historical matching success rate are obtained; and the dynamic thresholds are periodically adjusted based on the overall supply and demand volatility of the industry and the historical matching success rate.

[0039] After establishing the identification region based on the deviation category, the process also includes: extracting resource idle regions; obtaining the corresponding current agent based on the resource idle regions and using the obtained current agent as the first agent; obtaining the corresponding comprehensive competitiveness index based on the first agent and using the comprehensive competitiveness index as the first comprehensive competitiveness index; sorting the first comprehensive competitiveness index in descending order and generating a first candidate agent list; calculating the ratio between the product forecast demand and export supply capacity of the market gap regions; generating a gap-filling priority list based on the ratio; and associating the first candidate agent list and the gap-filling priority list to generate a resource pre-allocation plan.

[0040] To calculate the overall supply and demand volatility of the industry, the average supply and demand deviation can be calculated using the supply and demand deviation value. The standard deviation can then be calculated based on the average supply and demand deviation value. Dividing the standard deviation by the average supply and demand deviation value yields the overall supply and demand volatility of the industry. For the historical matching success rate, the electronic device acquires the number of successfully matched first regions within a second preset time period, and simultaneously acquires the number of matched second regions within the corresponding second preset time period. Dividing the number of first regions by the number of second regions yields the historical matching success rate.

[0041] The specific calculation formula for adjusting the dynamic threshold is as follows: ; in, The adjusted dynamic threshold; The dynamic threshold before adjustment; The learning rate; Target success rate; Historical match success rate; This is a correction factor.

[0042] Specifically, the process involves matching supply and demand gap heatmaps and candidate agent information to generate an expanded recommendation list, including: constructing a candidate agent pool based on candidate agent information; extracting corresponding business characteristics based on candidate agent information; obtaining a collaborative filtering model; inputting the business characteristics into the collaborative filtering model to calculate the similarity between candidate agents and market gap areas; ranking candidate agents based on similarity and generating expanded recommendations; generating recommendation reason text based on the expanded recommendations; and the recommendation reason text includes the association information between candidate agents and market gap areas.

[0043] The electronic device constructs a candidate agent pool and then extracts the business characteristics corresponding to the candidate agents. The dimensions of the business characteristics include, but are not limited to, export destination country vectors, product category preferences, and export scale levels. The collaborative filtering model can use the cosine similarity algorithm. After calculating the similarity, the electronic device ranks the candidate agents based on the similarity and generates extended recommendations, while also generating recommendation reason text, enabling users to better select suitable candidate agents.

[0044] As an optional implementation of this application, the method further includes: extracting data within a first preset period based on historical export data and using the extracted data as first calculation data; performing calculations based on the first calculation data to obtain a matching quality score; determining whether the matching quality score is less than a preset matching threshold; and if the matching quality score is less than the preset matching threshold, recording the current matching data and outputting feedback adjustment data.

[0045] The specific calculation process is as follows: The formula for calculating the quality score is: ; in, To match the quality score; This represents the actual export value. This represents the expected export value. Market share change rate; Product sales rate; Return rate; , , , All are preset weights and are known values.

[0046] As an optional implementation of this application, the method further includes: dynamically maintaining a sliding time window of a preset length based on the current time; collecting customs export data within the sliding time window; calculating the distribution change between the current multidimensional agent capability profile and the previous generation multidimensional agent capability profile at first preset intervals; calculating the data drift based on the distribution change; determining whether the data drift is greater than a preset drift threshold; and outputting adjustment parameters based on the data drift when the data drift is greater than the preset drift threshold.

[0047] Figure 2 A structural block diagram of a resource integration device 1000 based on customs data provided in this application embodiment is shown below. Figure 2 As shown, the resource integration device 1000 based on customs data includes: The cleaning module 1001 is used to obtain historical export data corresponding to each current agent based on customs data and clean the historical export data to obtain the first export analysis data. The first extraction module 1002 is used to extract data features based on the first export analysis data; the data features include at least the distribution of export destination countries, the proportion of export volume, and the main product categories. A calculation module 1003 is constructed to build a multi-dimensional agent capability profile based on data features and to calculate the comprehensive competitiveness index corresponding to each current agent entity based on the multi-dimensional agent capability profile. The first output module 1004 is used to output a resource allocation scheme based on the comprehensive competitiveness index; The first acquisition module 1005 is used to acquire import demand data from the target market; The first module 1006 is used to establish identification regions based on multi-dimensional agent capability profiles and target market import demand data; the identification regions include resource-idle regions, market gap regions, and high-potential regions. The first generation module 1007 is used to generate a heat map of supply and demand gaps based on the identified regions; The second acquisition module 1008 is used to acquire candidate agent entity information; The second generation module 1009 is used to generate an expanded recommendation list by matching the supply and demand gap heat map and candidate agent information.

[0048] Specifically, the construction of computing module 1003 includes: The first calculation submodule is used to calculate the regional coverage entropy value, market share dynamic value, and the fit score between the current agent and the product category; The second calculation submodule is used to calculate the average fitness score based on the fitness score. The first processing submodule is used to normalize the regional coverage entropy value, market share dynamic value and average fit score. The third calculation submodule is used to calculate the comprehensive competitiveness index based on the normalized regional coverage entropy value, market share dynamic value, and average fit score.

[0049] Specifically, the first module 1006 includes: The first extraction submodule is used to extract the export supply capacity of each current agent to each target market based on the multi-dimensional agent capability profile. The first acquisition submodule is used to acquire historical import demand data and macroeconomic indicators of the target market; The second acquisition submodule is used to acquire time series prediction models; The first estimation submodule is used to input historical import demand data and macroeconomic indicators into the time series forecasting model to estimate the product forecast demand in the target market within a future time window. The fourth calculation submodule is used to calculate the supply-demand deviation value based on export supply capacity and product forecast demand. The third acquisition submodule is used to acquire the deviation threshold; The first determination submodule is used to determine the deviation category of the target market based on the supply-demand deviation value and the deviation threshold. The first submodule is used to establish the recognition region based on the deviation category.

[0050] Specifically, the first determination submodule includes: The fourth acquisition submodule is used for deviation thresholds, which include multiple dynamic thresholds; the deviation thresholds include multiple dynamic thresholds. The first adjustment submodule is used to periodically adjust the dynamic threshold based on the overall supply and demand volatility of the industry and the historical matching success rate. The second extraction submodule is used to extract the idle resource area after the recognition area is established based on the deviation category; The fifth acquisition submodule is used to acquire the corresponding current agent based on the resource idle area and use the acquired current agent as the first agent; The sixth acquisition submodule is used to acquire the corresponding comprehensive competitiveness index based on the first agent and use the comprehensive competitiveness index as the first comprehensive competitiveness index; The first sorting submodule is used to sort the first comprehensive competitiveness index in descending order and generate the first candidate agent list; The fifth calculation submodule is used to calculate the ratio between the product forecast demand and export supply capacity based on market gap areas; The first generation submodule is used to generate a priority list for filling gaps based on the ratio. The second generation submodule is used to associate the first candidate agent list and the gap-filling priority list to generate a resource pre-allocation scheme.

[0051] Specifically, the second generation module 1009 includes: The first construction submodule is used to build a candidate agent pool based on the candidate agent subject information; The third extraction submodule is used to extract corresponding business characteristics based on the candidate agent entity information; The seventh submodule is used to obtain the collaborative filtering model; The sixth calculation submodule is used to input business characteristics into the collaborative filtering model to calculate the similarity between candidate agents and market gaps; The second ranking submodule is used to rank candidate agent entities based on similarity and generate extended recommendations; The third generation submodule is used to generate recommendation reason text based on extended recommendations; the recommendation reason text includes the association information between the candidate agent and the blank area.

[0052] In this optional embodiment, the resource integration device 1000 based on customs data further includes: The seventh calculation submodule is used to calculate the matching quality score based on the first calculation data; The first judgment submodule is used to determine whether the matching quality score is less than the preset matching threshold; if the matching quality score is less than the preset matching threshold, the current matching data is recorded and feedback adjustment data is output.

[0053] In this optional embodiment, the resource integration device 1000 based on customs data further includes: The first maintenance submodule is used to dynamically maintain a sliding time window of a preset length based on the current time; The first data acquisition submodule is used to collect customs export data within a sliding time window. The eighth calculation submodule is used to calculate the distribution change between the current generation multidimensional agent capability profile and the previous generation multidimensional agent capability profile at a first preset time interval. The ninth calculation submodule is used to calculate the data drift based on the distribution change. The second judgment submodule is used to determine whether the data drift amount is greater than the preset drift threshold; when the data drift amount is greater than the preset drift threshold, the adjustment parameters are output based on the data drift amount.

[0054] Figure 3 This is a structural block diagram of an electronic device 1100 provided in an embodiment of this application. The electronic device 1100 can be a mobile phone, tablet computer, PC, server, or other similar device. Figure 3 As shown, the electronic device 1100 includes a memory 1101, a processor 1102, and a communication bus 1103; the memory and the processor 1102 are connected via the communication bus 1103. The memory 1101 stores a computer program that can be loaded by the processor 1102 and executed as described in the above embodiments, which is a resource integration method based on customs data.

[0055] The memory 1101 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 1101 may include a stored program area and a stored managed data area. The stored program area may store instructions for implementing an operating system, instructions for at least one function, and instructions for implementing the resource integration method based on customs data provided in the above embodiments, etc. The stored managed data area may store managed data involved in the resource integration method based on customs data provided in the above embodiments, etc.

[0056] Processor 1102 may include one or more processing cores. Processor 1102 executes instructions, programs, code sets, or instruction sets stored in memory 1101, calls managed data stored in memory 1101, performs various functions of this application, and processes managed data. Processor 1102 may be at least one of the following: Application-Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), Central Processing Unit (CPU), controller, microcontroller, and microprocessor. It is understood that for different devices, the electronic devices used to implement the functions of processor 1102 may be other types, and this application embodiment does not specifically limit the specific devices used.

[0057] The communication bus 1103 may include a path for transmitting information between the aforementioned components. The communication bus 1103 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 1103 can be divided into an address bus, a managed data bus, a control bus, etc. For ease of representation, Figure 3 The symbol is represented by a single double arrow, but this does not mean that there is only one bus or one type of bus.

[0058] This application provides a computer storage medium storing a computer program that can be loaded by a processor and executed as described in the above embodiments, a resource integration method based on customs data.

[0059] In this embodiment, the computer storage medium can be a tangible device that holds and stores instructions used by the instruction execution device. The computer storage medium can be, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination thereof. Specifically, the computer storage medium can be a portable computer disk, a hard disk, a USB flash drive, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), speaker random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory stick, floppy disk, optical disk, magnetic disk, mechanical encoding device, or any combination thereof.

[0060] The terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

Claims

1. A resource integration method based on customs data, characterized in that, include: Based on customs data, historical export data corresponding to each current agent entity is obtained and the historical export data is cleaned to obtain the first export analysis data. Data features are extracted based on the first export analysis data; The data features include at least the distribution of export destination countries, the proportion of export volume, and the main product categories; Based on the data features, a multi-dimensional agent capability profile is constructed, and based on the multi-dimensional agent capability profile, the comprehensive competitiveness index corresponding to each current agent entity is calculated. Resource allocation schemes are output based on the aforementioned comprehensive competitiveness index; Obtain import demand data from the target market; An identification region is established based on the multi-dimensional agent capability profile and the target market import demand data; The identified areas include areas with idle resources, market gaps, and high-potential areas. A supply-demand gap heatmap is generated based on the identified region; Obtain candidate agent entity information; An expanded recommendation list is generated by matching the supply and demand gap heat map with the candidate agent entity information.

2. The resource integration method based on customs data according to claim 1, characterized in that, Based on the data features, a multi-dimensional agent capability profile is constructed, and based on the multi-dimensional agent capability profile, a comprehensive competitiveness index corresponding to each current agent entity is calculated, including: Calculate the regional coverage entropy value, market share dynamic value, and the compatibility score between the current agent and the product category; Calculate the average fit score based on the fit score; The regional coverage entropy value, the dynamic value of market share, and the average fit score are normalized. The comprehensive competitiveness index is calculated based on the normalized regional coverage entropy, the dynamic value of market share, and the average fit score.

3. The resource integration method based on customs data according to claim 1, characterized in that, The establishment of the identification region based on the multi-dimensional agent capability profile and the target market import demand data includes: Based on the multi-dimensional agent capability profile, the export supply capability of each current agent entity to each target market is extracted; Obtain historical import demand data and macroeconomic indicators for the target market; Obtain a time series forecasting model; The historical import demand data and the macroeconomic indicators are input into the time series forecasting model to estimate the projected product demand in the target market within a future time window. The supply-demand deviation value is calculated based on the export supply capacity and the projected demand for the product. Obtain the deviation threshold; The deviation category of the target market is determined based on the supply-demand deviation value and the deviation threshold. The identification region is established based on the deviation category.

4. The resource integration method based on customs data according to claim 3, characterized in that, The step of determining the deviation category of the target market based on the supply-demand deviation value and the deviation threshold includes: Obtain the overall supply and demand volatility of the industry and the historical matching success rate; The deviation threshold includes multiple dynamic thresholds; The dynamic threshold is periodically adjusted based on the overall supply and demand volatility of the industry and the historical matching success rate. After establishing the recognition region based on the deviation category, the method further includes: Extract the idle resource area; Based on the resource idle area, obtain the corresponding current agent and use the obtained current agent as the first agent; Based on the first agent, the corresponding comprehensive competitiveness index is obtained and the comprehensive competitiveness index is used as the first comprehensive competitiveness index; Sort the first comprehensive competitiveness index in descending order and generate a first candidate agent list; The ratio is calculated based on the projected demand for the product in the market gap area and the export supply capacity. A priority list for filling gaps is generated based on the ratio; The first candidate agent list and the gap-filling priority list are associated to generate a resource pre-allocation scheme.

5. A resource integration method based on customs data according to claim 1 or 4, characterized in that, The step of matching the supply-demand gap heatmap and the candidate agent information to generate an expanded recommendation list includes: A candidate agent pool is constructed based on the candidate agent entity information; Based on the candidate agent entity information, extract the corresponding business characteristics; Obtain the collaborative filtering model; The operational characteristics are input into the collaborative filtering model to calculate the similarity between the candidate agent and the market gap area; The candidate agent entities are ranked based on the similarity and extended recommendations are generated. Based on the extended recommendation, generate recommendation reason text; The recommendation reason text includes the association information between the candidate agent and the blank area.

6. The resource integration method based on customs data according to claim 1, characterized in that, The method further includes: Data within a first preset period is extracted based on the historical export data, and the extracted data is used as the first calculation data; A matching quality score is obtained by calculating based on the first calculation data; Determine whether the matching quality score is less than a preset matching threshold; If the matching quality score is less than the preset matching threshold, the current matching data is recorded and feedback adjustment data is output.

7. The resource integration method based on customs data according to claim 1, characterized in that, The method further includes: A sliding time window of a preset length is dynamically maintained based on the current time; Collect customs export data within the sliding time window; Calculate the distribution change between the current generation of the multidimensional agent capability profile and the previous generation of the multidimensional agent capability profile corresponding to the current agent subject at a first preset time interval; Calculate the data drift based on the aforementioned distribution change. Determine whether the data drift amount is greater than a preset drift threshold; When the data drift amount is greater than the preset drift threshold, an adjustment parameter is output based on the data drift amount.

8. A resource integration device based on customs data, characterized in that, include: The cleaning module is used to obtain historical export data corresponding to each current agent entity based on customs data and clean the historical export data to obtain the first export analysis data. The first extraction module is used to extract data features based on the first export analysis data; the data features include at least the distribution of export destination countries, the proportion of export volume, and the main product categories. A calculation module is constructed to build a multi-dimensional agent capability profile based on the data features and to calculate the comprehensive competitiveness index corresponding to each current agent entity based on the multi-dimensional agent capability profile. The first output module is used to output a resource allocation scheme based on the comprehensive competitiveness index; The first acquisition module is used to acquire import demand data from the target market. The first module is used to establish an identification region based on the multi-dimensional agent capability profile and the target market import demand data; the identification region includes resource idle regions, market gap regions, and high-potential regions. The first generation module is used to generate a supply and demand gap heat map based on the identified area; The second acquisition module is used to acquire candidate agent entity information; The second generation module is used to match the supply and demand gap heat map and the candidate agent entity information to generate an expanded recommendation list.

9. An electronic device, characterized in that, The device includes a processor coupled to a memory; the processor is configured to execute a computer program stored in the memory to cause the electronic device to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It includes a computer program or instructions that, when run on a computer, cause the computer to perform the method as described in any one of claims 1 to 7.

Images