A simulation analysis method and visualization system for goods trade data based on data correlation optimization
Through the method based on data correlation optimization, the iterative process of force-oriented layout algorithms is simplified, and the problems of slow speed and poor performance of traditional algorithms in the processing of global goods trade data are solved, achieving faster and more accurate complex network simulation analysis.
Patent Information
- Application Number
- CN202411124444.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-16
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-08-16
AI Technical Summary
Traditional force-oriented layout algorithms use slow processing speed and poor performance on global goods trade data when building complex networks, making it difficult to effectively analyze large-scale data sets, resulting in long calculation time and poor graphics results.
Using the simulation and analysis method of goods trade data based on data correlation optimization, the initial undirected graph is constructed by collecting and cleaning goods trade data, and using the maximum spanning tree algorithm and force-oriented layout algorithm for iterative optimization, simplifying the iteration process and extracting topological relationships , calculate the forces of nodes and edges in stages to improve calculation speed and accuracy.
It significantly improves the computing speed and accuracy of complex network simulations, reduces the computing complexity of force-oriented layout, improves the reliability of analysis results, and helps users better conduct correlation analysis of import and export trade data.
Smart Images

Figure CN118644346B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a data processing system or method specifically suitable for commercial, financial, management, supervision or forecasting purposes, and in particular to a goods trade data simulation analysis method and a visualization system based on data correlation optimization. Background Art
[0002] Graph visualization is the process of using computer graphics and image processing technology to convert data into graphics corresponding to the relationship between data and then display them in a computer system through program drawing. Organizing data and presenting it through a visualization system can assist users in analyzing and mining the value of data. Using complex networks to simulate and visualize large amounts of data, a common method is to use force-directed layout to achieve data clustering and relationship display. Global merchandise trade data involves multi-dimensional statistical results from a large number of countries and regions. Using complex network simulation to analyze the historical trade connections, upstream and downstream supply chain conditions, and changes in export industry structure behind them is of great significance to national and corporate decision-making.
[0003] However, the traditional force-directed iterative algorithm has the following problems when constructing complex networks to simulate global goods trade data:
[0004] Slow speed: Traditional force-directed layout is an algorithm based on the iteration of all node displacements. Each iteration requires recalculating the positions of all nodes. As the number of nodes increases, the amount of calculation also increases, resulting in a slow speed. The data on countries and regions involved in global trade and specific transaction categories are extremely large and complex, and it is almost impossible to calculate directly using raw data.
[0005] Poor performance: The computational complexity of traditional force-directed layout is high. For large-scale data sets, the algorithm takes a very long time to calculate. In addition, the initial position is random, and the iteration speed is difficult to control. After multiple iterations, it is difficult to ensure a good graphics effect, and it is difficult to obtain effective conclusions for complex network analysis. Summary of the invention
[0006] In view of this, the present invention proposes a commodity trade data simulation and analysis method and a visualization system based on data correlation optimization, which solves the problems of slow speed and poor performance when the traditional force-directed iterative algorithm is used to construct a complex network to simulate global commodity trade data.
[0007] The technical means adopted by the present invention are as follows:
[0008] In one aspect, the present invention provides a method for simulating and analyzing goods trade data based on data correlation optimization, the method comprising the following steps:
[0009] S1. Collect goods trade and macro-statistical data and establish a statistical database; the goods trade and macro-statistical data include: foreign trade enterprise statistics, customs transaction bills of lading data and macro-statistical data;
[0010] S2. Clean the data in the statistical database and add data labels according to the exporting economies, customs codes corresponding to the products, and years corresponding to the data; calculate the export correlation between the products corresponding to each customs code and the global import and export trade volume corresponding to each data label on an annual basis;
[0011] S3. Use the force-directed layout algorithm to simulate and calculate the export product association of economies through undirected graphs in complex networks, including:
[0012] S31. Based on the export correlation between products corresponding to each customs code calculated in units of years and the global import and export trade volume corresponding to each data tag, an initial undirected graph is constructed, wherein the undirected graph has all the export products corresponding to the four-digit customs codes with import and export trade in that year as vertices, and the correlation between the export products as edges, and the value of the edge is determined by the export correlation of the corresponding product;
[0013] S32. Using the maximum spanning tree algorithm, traverse from any vertex in the initial undirected graph to obtain a minimal connected subgraph that contains all vertices in the graph and has the strongest edge association; using the points and edges in the subgraph as the initial nodes and edges of the force-directed layout, using the simulated spring charge model, adding a repulsive force between every two nodes and an attractive force between the two nodes of each edge, the repulsive force is determined by the Euclidean distance between the nodes, and the attractive force is determined by the export association; using a random seed to generate the initial position, iterating through the force-directed layout algorithm, recording the coordinates of all points, the repulsive force between the points, and the attractive force between the corresponding vertices of the edges after each iteration; repeating the iterative process until the iteration stop condition is reached;
[0014] S33, for each node, only retain the edge with the highest export correlation and the edge weight value exceeding the preset value of the edge weight connected to this node in the original undirected graph, and only consider the gravitational force of the retained edge when performing force-directed layout calculation, and the repulsive force between all nodes is determined by the distance; for the new subgraph, use the coordinates of each node obtained in S32 as the initial position, iterate through the force-directed layout algorithm, and record the coordinates of all points, the repulsive force between all points, and the attractive force between all edges after each iteration; repeat the iterative process until the iteration stop condition is reached;
[0015] S34. For the initial undirected graph of S31, the coordinates of each node obtained in S33 are used as the initial positions, and the force-directed layout algorithm is used for iteration. After each iteration, the coordinates of all points, the repulsive forces based on distance of all points, and the attractive forces between the corresponding nodes of all edges are recorded. The aforementioned iterative process is repeated, and the coordinates of each node when the iteration is stopped are recorded. The coordinates of each node calculated at this time are the coordinates of each point of the undirected graph required for the complex network and the front-end display of the visualization system.
[0016] Furthermore, it also includes: S4, visually outputting the coordinates of each point of the undirected graph obtained in S34.
[0017] Furthermore, the data in the statistical database is cleaned, including:
[0018] Convert data from different sources into a standard format;
[0019] Cooperate with manual verification to verify the reliability of data and clean up duplicate and invalid data.
[0020] Furthermore, the export correlation between products corresponding to each customs code and the global import and export trade volume corresponding to each data tag in the database are calculated on an annual basis, including:
[0021] S21: Use structured query language to traverse the data of the specified year in the international goods trade database to obtain a list of countries and products involved in international trade imports and exports; the international goods trade database is a standardized universal dollar amount and customs code data item, which is comparable between countries;
[0022] S22: Collect the public data of customs classification rules of various economies, combine with manual verification, and remove all products from the list obtained in step 21 that are subject to large-scale embargoes, restrictions, or cannot be exported due to international treaties, or products that cannot be compared in terms of industrial capacity, to obtain the calculation target, and then use it to obtain a product list for competitiveness and export revenue analysis;
[0023] S23: Check whether all data in the database exist in both exporting and importing countries. For data that only exist in one of the exporting and importing countries, generate corresponding goods trade records in the corresponding exporting / importing countries to ensure that all trade records in the database exist in both the exporting and importing countries, and to ensure the accuracy of calculating the total import and export volume of each country or economy in the year;
[0024] S24: Through structured query language, based on the list of economies participating in international trade in the year obtained from S21 and the list of valid trade products obtained from S22, use all the data for the year obtained in S23 to calculate the total import and export volume of each country, the global total of each product, the global total volume of all trade and the percentage composition, obtain the required basic indicators, and generate the current calculation basic data table.
[0025] Furthermore, the indicator calculation includes:
[0026] The calculation formula for the explicit comparative advantage (RCA) of a country in evaluating the export competitiveness of a specific product in a given year is: the quotient of the proportion of the industry's exports in the country's total exports and the proportion of the industry's exports in the world's global exports:
[0027] ;
[0028] in, For an exporting country Export products export volume;
[0029] Combined with the calculated explicit comparative advantages of each economy’s export products in that year, the two products , The correlation coefficient The calculation method is: find all the exporting countries that export both products at the same time, and then calculate the conditional probability that a country that exports one of the products also exports the other product. The smaller one is the correlation coefficient between the two products:
[0030] ;
[0031] in, Represents the current analysis product The correlation coefficient of Representing the exporting country while also in the product and The export revealed comparative advantage coefficient is greater than 1. Products representing exporting countries Events where the export revealed comparative advantage coefficient is greater than 1, Products representing exporting countries Events where the export revealed comparative advantage coefficient is greater than 1, It is a conditional probability defined mathematically, and min represents the smaller value of the two conditional probabilities.
[0032] Furthermore, by associating goods trade data, the iteration process is simplified, the speed of complex network simulation calculation is accelerated, and the iteration stop conditions are achieved more quickly, including: the network reaches a stable state, the number of iterations reaches a preset iteration number threshold, or the iteration time reaches a preset iteration time threshold;
[0033] Starting from the preset number of iterations, the sum of the changes in the Euclidean distances of all vertex coordinates between two iterations is compared. When the sum of the displacements of all coordinates in the graph is less than the preset value, the complex network is considered to have iterated to a stable state.
[0034] In another aspect, the present invention further provides a visualization system for simulation analysis of goods trade data based on data correlation optimization, the visualization system comprising:
[0035] Interaction module, data processing module and visualization module;
[0036] The interactive module is used for the interaction between the user and the front-end graphical interface or console, and receives the user's input on the front-end graphical interface or console;
[0037] The data processing module executes the above-mentioned simulation analysis method of goods trade data based on data correlation optimization to obtain the coordinates of each point of the undirected graph required for the complex network and the front-end display of the visualization system;
[0038] The visualization module retrieves the standardized statistical data of global merchandise trade with corresponding data tags according to the user's input in the front-end graphical interface or console, and uses the node coordinates calculated by the data processing module to draw a complex network of export product associations, showing the export structure of the economy to be analyzed and the degree of association between the export products, as well as the related products of each export product.
[0039] Furthermore, in the interactive module, users can define the export economies, product ranges and time ranges they want to observe and study, select economies other than the current analysis when generating simulation results, and compare the trade surpluses and deficits of import and export products and the competition status of global markets based on international trade statistics big data, and analyze the trade competition and complementarity of import and export structures.
[0040] Compared with the prior art, the present invention has the following advantages:
[0041] The present invention proposes a simulation analysis method and visualization system for goods trade data based on data correlation optimization, which uses the correlation between goods trade data to extract its topological relationship, generates a simple subgraph of the complex network to be analyzed, and implements force-guided layout of complex network iteration and related parameter calculation in stages, thereby increasing the calculation speed and improving the accuracy of complex network simulation. This method can reduce the computational complexity of force-guided layout to construct complex networks and reduce the time required for iteration, improve the reliability of analysis results, and help users better perform complex network simulation of import and export trade data and goods import and export correlation analysis. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0043] Figure 1 It is a flowchart of a method for simulating and analyzing goods trade data based on data correlation optimization in an embodiment of the present invention;
[0044] Figure 2 It is a flowchart of a complex network simulation method based on trade data relationship optimization force-directed layout in an embodiment of the present invention. DETAILED DESCRIPTION
[0045] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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 should fall within the scope of protection of the present invention.
[0046] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0047] The present invention proposes a simulation analysis method and visualization system for goods trade data based on data correlation optimization, which realizes accurate classification and processing of goods trade data by adopting separate or integrated modules, obtains advantageous export industries of various countries by calculation and quantitatively analyzes the export correlation between products, and then determines reliable force-guided layout algorithm iteration starting parameters and clustering division groups through the topological relationship of product export correlation, thereby improving the visualization iteration speed and visualization analysis effect of complex network of trade data, and providing more accurate and sophisticated data and visualization system support for users' analysis and decision-making.
[0048] like Figure 1 As shown, a method for simulating and analyzing goods trade data based on data correlation optimization in an embodiment of the present invention specifically includes the following steps:
[0049] S1. Collect goods trade and macro-statistics data and establish a statistical database;
[0050] Among them, goods trade and macro-statistical data include: foreign trade enterprise statistics, customs transaction bill of lading data and macro-statistical data.
[0051] S2. Clean the data in the statistical database, save the cleaned data in the database, and add labels according to the exporting economy, the customs code corresponding to the product, and the year corresponding to the data; calculate the export correlation between the products corresponding to each customs code and the global import and export trade volume corresponding to each data label in the database on an annual basis.
[0052] In the specific implementation, the data is processed according to the following steps:
[0053] S21: Use structured query language to traverse the data of the specified year in the international goods trade database to obtain a list of countries and products involved in international trade imports and exports; the international goods trade database is a standardized universal dollar amount and customs code data item, which is comparable between countries;
[0054] S22: Collect the public data of customs classification rules of various economies, combine with manual verification, and remove all products from the list obtained in step 11 that are subject to large-scale bans, restrictions, or cannot be exported due to international treaties, or products that cannot be compared in terms of industrial capacity, to obtain the calculation target, and then use it to calculate the product list for competitiveness and export revenue analysis;
[0055] S23: Check whether all data in the database exist in both exporting and importing countries. For data that only exist in one of the exporting and importing countries, generate corresponding goods trade records in the corresponding exporting / importing countries to ensure that all trade records in the database exist in both the exporting and importing countries, and to ensure the accuracy of calculating the total import and export volume of each country or economy in the year;
[0056] S24: Through structured query language, based on the list of economies participating in international trade in the year obtained from S21 and the list of valid trade products obtained from S22, use all the data for the year obtained in S23 to calculate the total import and export volume of each country, the global total of each product, the global total volume of all trade and the percentage composition, obtain the required basic indicators, and generate the current calculation basic data table.
[0057] The indicator calculation specifically includes:
[0058] The formula for calculating the revealed comparative advantage (RCA) of a country in evaluating the export competitiveness of a specific product in a given year is the quotient of the proportion of the industry's exports in the country's total exports and the proportion of the industry's exports in the world's global exports:
[0059] ;
[0060] in, For an exporting country Export products export volume;
[0061] Combined with the calculated table of explicit comparative advantage index of export products of each economy in that year, for two products , , the correlation coefficient of these two products The calculation method is: find all exporting countries that export both products (RCA>1); then calculate the conditional probability that a country that exports one product also exports the other product, and the smaller one is the correlation coefficient between the two products:
[0062] ;
[0063] in, Represents the current analysis product The correlation coefficient of Representing the exporting country while also in the product and The export revealed comparative advantage coefficient is greater than 1. Products representing exporting countries Events where the export revealed comparative advantage coefficient is greater than 1, Products representing exporting countries Events where the export revealed comparative advantage coefficient is greater than 1, It is a conditional probability defined mathematically, and min represents the smaller value of the two conditional probabilities.
[0064] S3. Use the force-directed layout algorithm to simulate and analyze the export product association of economies through the undirected graph model in complex networks;
[0065] The core idea of force-directed layout is to simulate the interaction between each node (vertices in the graph) and between the node and the boundary of the layout space. These forces mainly include: Repulsion: simulates the repulsive force between positively charged particles to prevent nodes from being too close and make the graph layout more uniform. Gravity: simulates the effect of springs to keep connected nodes close to each other and maintain the connectivity of the graph. Boundary constraint: restricts nodes within the layout space to prevent nodes from floating out of the visible area.
[0066] The force-directed layout algorithm process includes: First, all nodes are randomly placed in the layout space. According to the distance and connection relationship between nodes, the repulsive and attractive forces between each pair of nodes, as well as the forces between nodes and boundaries are calculated. Based on the calculated forces, the position of each node is updated. The steps of calculating forces and moving nodes are repeated until the system reaches a steady state, that is, all forces are balanced or the changes are very small. When the layout is stable or the preset number of iterations is reached, the algorithm ends.
[0067] In this embodiment, the traditional force-directed layout algorithm is improved. The correlation between goods trade data is used to extract its topological relationship, generate a simple subgraph of the complex network to be analyzed, and implement the force-guided layout complex network iteration and related parameter calculation in stages. This can reduce the computational complexity of force-directed layout to construct complex networks and reduce the time required for iteration, thereby improving the reliability of analysis results and helping users to better perform complex network simulation of import and export trade data and goods import and export correlation analysis.
[0068] In specific implementation, Figure 2 As shown, S3 includes the following steps:
[0069] S31. Construction of initial undirected graph: An undirected graph is used as a simulation analysis model. The four-digit customs codes corresponding to all export products in the year are used as the vertices of the undirected graph. The edges of the undirected graph are the associations between export products, and the values of the edges are determined by the export association of the corresponding products.
[0070] S32. Generate associated subgraphs: Using the maximum spanning tree algorithm, traverse from any vertex in the initial undirected graph to obtain a minimal connected subgraph that contains all vertices in the graph and has the strongest edge association. Use the points and edges in the minimal connected subgraph as the initial points and edges of the force-directed layout. Here, the simulated spring charge model is used (in the force-directed layout algorithm, the simulated spring charge model is a technology used for graph drawing that uses physical concepts to optimize and improve the layout of the graph. This model combines the repulsion between charges (similar to Coulomb repulsion) and the attraction between connected nodes (similar to spring force) to determine the position of the node on the canvas). Add a repulsive force between every two nodes and an attractive force between the two nodes of each edge. The repulsive force is determined by the Euclidean distance between the nodes, and the attractive force is determined by the export association. Use a random seed to generate the initial position, iterate through the force-directed layout algorithm, and record the coordinates of all points, the repulsive force between the points, and the attractive force between the corresponding vertices of the edges after each iteration. Repeat the above iterative process, and compare the sum of the changes in the Euclidean distance of all coordinates between two iterations starting from the preset number of iterations. When the sum of the displacements of all coordinates in the graph is less than the preset value, it is considered that the complex iteration has reached a stable state. The iteration ends when one of the following three situations occurs: when the network reaches a stable state, the number of iterations reaches a preset iteration number threshold, or the iteration time reaches a preset iteration time threshold, the iteration is stopped and the coordinates of each node at this time are recorded.
[0071] S33. Speed up the operation by controlling the average degree of nodes in the complex network:
[0072] Based on the analysis of the calculation results of the goods trade association, for each node, only the edges with the highest export association and edge weight value (export association) exceeding the preset value of the edge weight connected to this node in the original undirected graph (the preset value of edge weight is preferably 99%) are retained (product pairs that do not meet this standard have no statistically significant export association). When performing force-directed layout calculations, only the gravitational force of these edges is considered, and the repulsive force between all nodes is determined by the distance. For the new subgraph, the coordinates of each node calculated in S32 are used as the initial position, and the force-directed layout algorithm is iterated. After each iteration, the coordinates of all points, the repulsive force between all points, and the attractive force between all edges are recorded. Repeat the above iterative process, and compare the sum of the changes in the Euclidean distance of all coordinates between two iterations starting from the preset number of iterations. When the sum of the displacements of all coordinates in the graph is less than the preset value, it is considered that the complex iteration has reached a stable state, and the iteration ends when one of the following three situations occurs: when the network reaches a stable state, the number of iterations reaches the preset number of iterations threshold, or the iteration time reaches the preset iteration time threshold, the iteration is stopped, and the coordinates of each node at this time are recorded.
[0073] S34, for the initial undirected graph of S31, the coordinates calculated by S33 are used as the initial node positions, and the force-directed layout algorithm is used for iteration. After each iteration, the coordinates of all points, the repulsive forces based on distance of all points, and the attractive forces between all edges corresponding to the nodes are recorded. Repeat the above-mentioned iterative process, and compare the sum of the displacements of all coordinates between two iterations starting from the preset number of iterations. When the sum of the displacements of all coordinates in the graph is less than the preset value, it is considered that the iteration has reached a stable state. Repeat this process until one of the following three situations occurs: when the network reaches a stable state, the number of iterations reaches a preset iteration number threshold, or the iteration time reaches a preset iteration time threshold, the iteration is stopped, and the coordinates of each node at this time are recorded. At this time, the coordinates of each node calculated are the coordinates of each point in the undirected graph required for the complex network and visualization system front-end display.
[0074] Based on the goods trade data simulation analysis method based on data relevance optimization in the above embodiment, a goods trade data simulation analysis visualization system based on data relevance optimization is constructed in this embodiment. The visualization system includes:
[0075] Interaction module, data processing module and visualization module.
[0076] Among them, the interactive module is used for the interaction between the user and the front-end graphical interface or console, and receives the user's input in the front-end graphical interface or console. In the interactive module, the user can also define the export economy, product range and time range that he wants to observe and study, select non-current analysis economies in the generated simulation results, and compare the trade surplus and deficit of import and export products and the competition status of various global markets based on the big data of international trade statistics, and analyze the trade competition and complementarity of import and export structures.
[0077] The data processing module executes a simulation analysis method for goods trade data based on data correlation optimization in the above embodiment to obtain the coordinates of each point in the undirected graph required for the complex network and visualization system front-end display.
[0078] The visualization module retrieves the standardized statistical data of global merchandise trade with corresponding data tags according to the user's input in the front-end graphical interface or console, and uses the node coordinates calculated by the data processing module to draw a complex network of export product associations, showing the export structure of the economy to be analyzed and the degree of association between the export products, as well as the related products of each export product.
[0079] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A simulation analysis method for goods trade data based on data correlation optimization, characterized in that: The method comprises the following steps: S1. Collect goods trade and macro-statistical data and establish a statistical database; the goods trade and macro-statistical data include: foreign trade enterprise statistics, customs transaction bills of lading data and macro-statistical data; S2. Clean the data in the statistical database and add data labels according to the exporting economy, the customs code corresponding to the product, and the year corresponding to the data; calculate the export correlation between the products corresponding to each customs code and the global import and export trade volume corresponding to each data label on an annual basis, including: S21: Use structured query language to traverse the data of the specified year in the international goods trade database to obtain a list of countries and products involved in international trade imports and exports; the international goods trade database is a standardized universal dollar amount and customs code data item, which is comparable between countries; S22: Collect the public data of customs classification rules of various economies, combine with manual verification, and remove all products from the list obtained in step 21 that are subject to large-scale embargoes, restrictions, or cannot be exported due to international treaties, or products that cannot be compared in terms of industrial capacity, to obtain the calculation target, and then use it to obtain a product list for competitiveness and export revenue analysis; S23: Check whether all data in the database exist in both exporting and importing countries. For data that only exist in one of the exporting and importing countries, generate corresponding goods trade records in the corresponding exporting / importing countries to ensure that all trade records in the database exist in both the exporting and importing countries, and to ensure the accuracy of calculating the total import and export volume of each country or economy in the year; S24: Using structured query language, based on the list of economies participating in international trade in the current year obtained in S21 and the list of valid trade products obtained in S22, use all the data for the current year obtained in S23 to calculate the total import and export volume of each country, the global total volume of each product, the global total volume of all trade, and the percentage composition, obtain the required basic indicators, and generate the current calculation basic data table; Among them, the indicator calculation includes: The calculation formula for the explicit comparative advantage (RCA) of a country in evaluating the export competitiveness of a specific product in a given year is: the quotient of the proportion of the industry's exports in the country's total exports and the proportion of the industry's exports in the world's global exports: ; in, For an exporting country Export products export volume; Combined with the calculated explicit comparative advantages of each economy’s export products in that year, the two products , The correlation coefficient The calculation method is: find all the exporting countries that export both products at the same time, and then calculate the conditional probability that a country that exports one of the products also exports the other product. The smaller one is the correlation coefficient between the two products: ; in, Represents the current analysis product The correlation coefficient of Representing the exporting country while also in the product and The export revealed comparative advantage coefficient is greater than 1. Products representing exporting countries Events where the export revealed comparative advantage coefficient is greater than 1, Products representing exporting countries Events where the export revealed comparative advantage coefficient is greater than 1, It is a conditional probability defined mathematically, and min represents the smaller value of the two conditional probabilities; S3. Use the force-directed layout algorithm to simulate and analyze the export product associations of economies through an undirected graph model in a complex network, including: S31. Based on the export correlation between products corresponding to each customs code calculated in units of years and the global import and export trade volume corresponding to each data tag, an initial undirected graph is constructed, wherein the undirected graph has all the export products corresponding to the four-digit customs codes with import and export trade in that year as vertices, and the correlation between the export products as edges, and the value of the edge is determined by the export correlation of the corresponding product; S32. Using the maximum spanning tree algorithm, traverse from any vertex in the initial undirected graph to obtain a minimal connected subgraph that contains all vertices in the graph and has the strongest edge association; using the points and edges in the minimal connected subgraph as the initial nodes and edges of the force-directed layout, using the simulated spring charge model, adding a repulsive force between every two nodes and an attractive force between the two nodes of each edge, the repulsive force is determined by the Euclidean distance between the nodes, and the attractive force is determined by the export association; using a random seed to generate the initial position, iterating through the force-directed layout algorithm, recording the coordinates of all points, the repulsive force between the points, and the attractive force between the corresponding vertices of the edges after each iteration; repeating the iterative process until the iteration stop condition is reached; S33, for each node, only retain the edge with the highest export correlation and the edge weight value exceeding the preset value of the edge weight connected to this node in the original undirected graph, and only consider the gravitational force of the retained edge when performing force-directed layout calculation, and the repulsive force between all nodes is determined by the distance; for the new subgraph, use the coordinates of each node obtained in S32 as the initial position, iterate through the force-directed layout algorithm, and record the coordinates of all points, the repulsive force between all points, and the attractive force between all edges after each iteration; repeat the iterative process until the iteration stop condition is reached; S34. For the initial undirected graph of S31, the coordinates of each node obtained in S33 are used as the initial positions, and the force-directed layout algorithm is used for iteration. After each iteration, the coordinates of all points, the repulsive forces based on distance of all points, and the attractive forces between the corresponding nodes of all edges are recorded. The aforementioned iterative process is repeated, and the coordinates of each node when the iteration is stopped are recorded. The coordinates of each node calculated at this time are the coordinates of each point of the undirected graph required for the complex network and the front-end display of the visualization system.
2. The method for simulating and analyzing goods trade data based on data correlation optimization according to claim 1 is characterized in that: Also includes: S4. Visualize and output the coordinates of each point in the undirected graph obtained in S34.
3. The method for simulating and analyzing goods trade data based on data correlation optimization according to claim 1 is characterized in that: Clean the data in the statistical database, including: Convert data from different sources into a standard format; Cooperate with manual verification to verify the reliability of data and clean up duplicate and invalid data.
4. The method for simulating and analyzing goods trade data based on data correlation optimization according to claim 1 is characterized in that: By associating goods trade data, the iteration process is simplified, the speed of complex network simulation calculation is accelerated, and the iteration stop conditions are achieved more quickly, including: the network reaches a stable state, the number of iterations reaches a preset iteration number threshold, or the iteration time reaches a preset iteration time threshold; Starting from the preset number of iterations, the sum of the changes in the Euclidean distances of all vertex coordinates between two iterations is compared. When the sum of the displacements of all coordinates in the graph is less than the preset value, the complex network is considered to have iterated to a stable state.
5. A visualization system for simulation analysis of goods trade data based on data correlation optimization, characterized in that: The visualization system comprises: Interaction module, data processing module and visualization module; The interactive module is used for the interaction between the user and the front-end graphical interface or console, and receives the user's input on the front-end graphical interface or console; The data processing module executes a simulation analysis method for goods trade data based on data correlation optimization as described in any one of claims 1 to 4 to obtain the coordinates of each point of the undirected graph required for the complex network and the front-end display of the visualization system; The visualization module retrieves the standardized statistical data of global merchandise trade with corresponding data tags according to the user's input in the front-end graphical interface or console, and uses the node coordinates calculated by the data processing module to draw a complex network of export product associations, showing the export structure of the economy to be analyzed and the degree of association between the export products, as well as the related products of each export product.
6. A visualization system for simulation analysis of goods trade data based on data correlation optimization according to claim 5, characterized in that: In the interactive module, users can define the export economies, product ranges and time ranges they want to observe and study, select economies other than the current analysis when generating simulation results, and compare the trade surpluses and deficits of import and export products and the competition status of global markets based on international trade statistics big data, as well as analyze the trade competition and complementarity of import and export structures.
Citation Information
Patent Citations
A global trade big data processing method
CN109918358A
Data relevance analysis method and system, and readable storage medium
CN110990777A
Trade index visualization system and method based on international trade commodity data
CN114547400A
Agricultural product trade information network service method
CN117853068A
Global trade transfer influence analysis method and visualization system
CN118154224A
Cited By
Trade rule reasoning method and system based on bidirectional mapping and storage medium
CN121189482A
A trade rule reasoning method and system based on bidirectional mapping and a storage medium
CN121189482B
