Cross-border market intelligent recommendation method and system based on multi-dimensional customs data fusion analysis

Abstract

The present application relates to data analysis technology, specifically to a cross-border market intelligent recommendation method and system based on multi-dimensional customs data fusion analysis, including a customs data analysis module and a competitive data analysis module; the present application classifies product information in a tree form, integrates multi-dimensional information such as import volume, trade flow, competitive data, and deeply mines market segmentation demand; and uses a linear regression model to predict import volume trend and time to reach the estimated maximum value, eliminates the interference of accidental factors, and ensures the stability of the recommended market demand; through analysis of trade correlation and product complementarity, enterprises can choose markets with strong product complementarity and little influence from competitors, reducing trade fluctuation risk; through quantifying the competitive difference between the enterprise and competitors; through a weighted scoring model, the target market is dynamically evaluated and sorted, and countries with high stability and strong competitiveness are preferentially recommended.

Description

Technical Field

[0001] This invention relates to data analysis technology, specifically to a method and system for intelligent recommendation of cross-border markets based on multi-dimensional customs data fusion analysis. Background Technology

[0002] In the field of cross-border trade, enterprises face core problems when expanding into international markets, such as inaccurate market demand analysis and insufficient assessment of the competitive environment. Existing methods cannot accurately analyze data volatility when faced with sudden changes in import volumes caused by short-term policy stimulus or accidental trade events, leading to misleading enterprises' judgments on market stability. At the same time, the lack of quantitative analysis of competitors makes it difficult for enterprises to assess their own competitiveness in target markets, potentially leading to blind entry into highly competitive markets and wasted resources.

[0003] To address the aforementioned technical problems, this application proposes a solution. Summary of the Invention

[0004] The purpose of this invention is to solve the problems raised in the background art, and to propose a cross-border market intelligent recommendation method and system based on multi-dimensional customs data fusion analysis.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] The intelligent recommendation method for cross-border markets based on multi-dimensional customs data fusion analysis includes the following steps:

[0007] S1: Classify your own products according to the internationally recognized HS code, establish a "product type hierarchy", and under the "product type hierarchy" establish a "product material hierarchy" by material, and then establish a "product specification hierarchy" by specification parameters. By determining whether the quantity of the parent hierarchy is equal to the total quantity of the child hierarchy, verify the matching of the quantities of each hierarchy to ensure the accuracy of the information.

[0008] S2: Obtain import volume data for each country within a set time period, calculate and compare the import volume difference between adjacent time points, and mark countries whose import demand is gradually increasing; analyze the historical import volume data of the marked countries, remove outliers, estimate the maximum value of the marked countries, and determine the time required to reach the maximum value based on the estimated maximum value, and determine whether the output to the marked countries is stable.

[0009] S3: Again, determine whether the relationship with the stable exporting country is stable by using trade correlation and import / export product complementarity index; screen out the countries determined to be stable exporting countries, analyze the market share, price competitiveness, and product structure matching degree of the screened countries, and combine the correlation coefficient and complementarity index to comprehensively evaluate the competitiveness of itself and its competitors in the market, screen the stable exporting countries again, and intelligently recommend the top-ranked stable exporting countries based on the ranking of the screening results.

[0010] As a preferred embodiment of the present invention, the intelligent recommendation system includes a customs data analysis module and a competition data analysis module;

[0011] The customs data analysis module establishes a tree diagram to classify products by type, material, and specifications; it assesses the demand of each country based on import volume data, filters countries with high demand, and estimates the maximum import volume of each selected country by removing outliers. This estimate is then compared with the country's import growth rate to preliminarily determine whether the country is a stable exporter; finally, it calculates the correlation coefficient and complementarity index between its own products and various countries during the trade process, further filtering the preliminarily identified stable exporters.

[0012] The competitive data analysis module analyzes the market share, price competitiveness, and product structure matching degree of stable exporting countries. It also retrieves the correlation coefficients and complementarity indices from the customs data analysis module to comprehensively evaluate its own competitiveness in the market and that of its competitors. The module further filters the stable exporting countries and intelligently recommends the top-ranked stable exporting countries based on the ranking of the filtering results.

[0013] In a preferred embodiment of the present invention, the steps for the customs data analysis module to establish the tree diagram are as follows:

[0014] K1: In accordance with the internationally accepted Harmonized System for Product Names and Coding, query and record the coding information of its own products, and then establish a corresponding number of tree diagram sub-levels based on the number of its own product coding information, and mark the sub-levels as product type levels;

[0015] K2: Obtain the material information of the product under the corresponding product code information, and build a corresponding number of tree diagram sub-levels under the product type level based on the number of obtained corresponding product material information, and mark the sub-level as the product material level.

[0016] K3: Obtain the product specifications under the corresponding product material information, and based on the number of corresponding product specifications obtained, create a corresponding number of tree diagram sub-levels under the product material level, and mark the sub-level as the product specification level.

[0017] K4: Count the number of items corresponding to each sub-level. If the number of items in the parent level is equal to the total number of items in the sub-levels, the record is considered correct; otherwise, the information of the parent level and all sub-levels under the parent level is recounted.

[0018] In a preferred embodiment of the present invention, the customs data analysis module analyzes the corresponding product demand as follows:

[0019] Q1: Retrieve the coding information of the product with the largest production volume, then obtain the import volume data of the corresponding HS code products in each country within a set time period from the customs data, sort the obtained import volume data of each country according to time, calculate the difference between the import volume data of adjacent time points, and compare the differences.

[0020] Q2: If the difference gradually increases over time, it is determined that the country's demand for the product is gradually increasing, and the country is marked as an exporter; if the difference gradually decreases over time, it is determined that the country's demand for the product is gradually decreasing; if the fluctuation of the difference over time is within the set fluctuation range, it is determined that the country's demand for the product tends to stabilize.

[0021] As a preferred embodiment of the present invention, the preliminary determination steps for stable export countries by the customs data analysis module are as follows:

[0022] P1: Obtain historical import data for countries with output markers, calculate the mean A1 and standard deviation B of the obtained historical import data, set the fluctuation range of the import data based on the calculated mean A1 and standard deviation B, and mark the import data outside the set fluctuation range as outliers.

[0023] P2: After removing outliers, calculate the mean A2 of the remaining import volume data. Use the calculated mean A2 as the historical average import volume data of the corresponding output label country. Calculate the difference between the mean A2 and the extreme value in the historical import volume data, and select the maximum difference as the estimated fluctuation value d. Then the estimated maximum import volume data of that country is A2+d.

[0024] P3: Obtain the import volume data of the corresponding products of the output marked country within a set time period, sort these import volume data according to time, and substitute the import volume data of the corresponding time data into the formula y=kx+b to obtain the specific values ​​of the coefficients k and b in the formula for the import volume variable data y and the time variable data x.

[0025] P4: Substituting the maximum value A2+d and the real-time import volume data of the output-marked country for the current time period into y=kx+b, we can deduce that after time x1, the import volume data of the corresponding product of the output-marked country will reach the estimated maximum value. If x1≥x STherefore, it is preliminarily determined that the country with the export mark can stably export the corresponding product. S The preset controllable time is used; otherwise, it is determined that the output of the corresponding product to the country marked with the output label is unstable.

[0026] In a preferred embodiment of the present invention, the customs data analysis module performs the following secondary determination step for stable export countries:

[0027] N1: Let X it To output the import volume of similar products of country i in period t, Y jt Let be the import volume of the j-th major trading country of the world in period t. Then, what is the correlation coefficient between this exporting country and the import volumes of all major trading countries? To output the average import volume of similar products in country i, Let be the average import volume of similar products in the j-th major trading country, and n be the number of time periods; average correlation coefficient. m represents the number of major trading partners;

[0028] N2: By obtaining the set of imported product information JK1 and the set of exported product information CK2 from the customs data of the output marked country, for product c, the export quantity of the product itself is CK. 2c The import quantity of the exporting country is JK. 1c Then the complementarity index CK2∩JK1 is the intersection of its own set of exported products CK2 and the set of imported products JK1 of the exporting country;

[0029] N3: Average correlation coefficient calculated using the maximum-minimum normalization method. Normalizing the complementarity index CI yields... and CI G Then the stability score e1 and e2 are the weight coefficients of the corresponding terms, respectively;

[0030] N4: Calculate stability scores for countries initially identified as stable outputs, and select those with stability scores WD > WD. max The output countries are labeled and sorted according to their scores. WD max The preset stability score threshold is used.

[0031] In a preferred embodiment of the present invention, the competitive data analysis module performs the following steps for analyzing competitiveness:

[0032] R1: For product c, its export quantity is CK. 2c The export volume of competitor m is CKm 3cThen, the market share of itself and its competitor D are respectively If FE2>max(FE3), then it is determined that the product c produced by itself has a dominant market share.

[0033] R2: The export transaction amount divided by the export quantity equals the unit price of the exported product. The export unit price of the product itself is DJ2, and the export unit price of the competitor is DJ3. If DJ2≤DJ3 and FE2 is on an upward trend, then the product itself is determined to be more competitive in the market.

[0034] R3: Overlap Index The higher the OI (Overlap Index), the better the match between the product structure and the target market demand, and the stronger the competitiveness.

[0035] R4: After normalization, we get FE 2G DJ 2G and OI G The market competitiveness index c1, c2, ω3, ω4, and ω5 are the weight coefficients of the corresponding items; then the market competitiveness index CS3 of the competitors is calculated. If CS2 > CS3, it is determined that the competitiveness of the company is better than that of the competitors.

[0036] R5: After obtaining the stability score, output the ranking of the marked countries. Among these marked countries, those with their own competitiveness are better than their competitors. Then, the competitiveness score is calculated and the marked countries are ranked according to the competitiveness score. Recommendations are made according to the ranking.

[0037] Compared with the prior art, the beneficial effects of the present invention are:

[0038] 1. By hierarchically dividing product information into tree structures and integrating multi-dimensional information such as import volume, trade flow, and competitive data, we can deeply explore segmented market demands; outliers are eliminated, and linear regression models are used to predict import volume trends and the time to reach the estimated maximum value, thus eliminating the interference of accidental factors and ensuring stable market demand for recommendations; through the analysis of trade relevance and product complementarity, enterprises can select markets that are highly complementary to their own products and less affected by competitors, thereby reducing the risk of trade fluctuations.

[0039] 2. By introducing indicators such as market share, price competitiveness, and overlap index, and combining them with correlation coefficient and complementarity index, the competitive differences between the company and its competitors are quantified; through a weighted scoring model, the target market is dynamically evaluated and ranked, and countries with high stability and strong competitiveness are given priority recommendations. Attached Figure Description

[0040] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0041] Figure 1 This is a system flowchart of the present invention. Detailed Implementation

[0042] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Example:

[0044] Please see Figure 1 As shown, a cross-border market intelligent recommendation method and system based on multi-dimensional customs data fusion analysis includes a customs data analysis module and a competitive data analysis module.

[0045] The customs data analysis module establishes a tree diagram to classify products by type, material, and specifications; it assesses the demand of each country based on import volume data, filters countries with high demand, and estimates the maximum import volume of each selected country by removing outliers. This estimate is then compared with the country's import growth rate to preliminarily determine whether the country is a stable exporter; finally, it calculates the correlation coefficient and complementarity index between its own products and various countries during the trade process, further filtering the preliminarily identified stable exporters.

[0046] A tree diagram is created to summarize the product information. The coding information of the products is queried and recorded according to the internationally recognized Harmonized System for Trade Names and Coding (HS). Then, a corresponding number of sub-levels are created based on the number of product coding information, and these sub-levels are marked as product type levels. Next, the material information of the products under the corresponding product coding information is obtained. Based on the number of obtained material information, a corresponding number of sub-levels are created under the product type level, and these sub-levels are marked as product material levels. Finally, the specifications of the products under the corresponding material information are obtained. Based on the number of obtained specifications, a corresponding number of sub-levels are created under the product material level, and these sub-levels are marked as product specification levels. The number of items in each sub-level is counted. If the number in the parent level equals the sum of the numbers in the sub-levels, the record is considered correct; otherwise, the information in the parent level and all its sub-levels is re-counted.

[0047] The system acquires production data for its own product type and sorts the corresponding product type's coding information from largest to smallest production data. It then retrieves the coding information for the product with the largest production data. Next, it obtains import volume data for the corresponding HS codes of various countries within a specified time period from customs data. The acquired import volume data for each country is sorted by time, and the difference between import volume data at adjacent time points is calculated and compared. If the difference gradually increases over time, it is determined that the country's demand for the product is gradually increasing, and that country is marked as such. If the difference gradually decreases over time, it is determined that the country's demand for the product is gradually decreasing. If the fluctuation of the difference over time is within a set fluctuation range, it is determined that the country's demand for the product is stabilizing.

[0048] Historical import volume data for countries with output markers is acquired. The mean A1 and standard deviation B of the acquired historical import volume data are calculated. The calculated mean A1 and standard deviation B are used to set the fluctuation range of the import volume data [A1-a*B, A1+a*B]. Import volume data outside the set fluctuation range are marked as outliers. The number of outliers is JS. Y Statistical analysis was conducted, and the results were compared with the total import volume data obtained (JS). Z If a comparison is made, If the fluctuation range is deemed too small, the preset multiple 'a' is reduced by one, and the ratio of the number of outliers to the total number within the fluctuation range is recalculated until the condition is met. Stop when k1 is the preset proportional coefficient; after removing outliers, calculate the mean A2 of the remaining import volume data. Use the calculated mean A2 as the historical average import volume data of the corresponding output marker country. Calculate the difference between the mean A2 and the extreme value in the historical import volume data, and select the maximum difference as the estimated fluctuation value d. Then the estimated maximum import volume data of the country is A2+d.

[0049] Obtain import volume data for the corresponding product from the output-marked country within a specified time period. Sort this import volume data by time, and substitute the import volume data for the corresponding time period into the formula y = kx + b to obtain the specific values ​​of the coefficients k and b in the formula for the import volume variable y and the time variable x. Substitute the maximum value A2 + d and the real-time import volume data of the output-marked country for the current time period into y = kx + b to determine that after time x1, the import volume data of the corresponding product of the output-marked country will reach the estimated maximum value. If x1 ≥ x S Therefore, it is preliminarily determined that the country with the export mark can stably export the corresponding product. S The preset controllable time is used; otherwise, it is determined that the output of the corresponding product to the country marked with the output label is unstable.

[0050] Let X itTo output the import volume of similar products of country i in period t, Y jt Let be the import volume of the j-th major trading country of the world in period t. Then, what is the correlation coefficient between this exporting country and the import volumes of all major trading countries? To output the average import volume of similar products in country i, Let be the average import volume of similar products in the j-th major trading country, and n be the number of time periods; average correlation coefficient. m represents the number of major trading countries, and the average correlation coefficient is... The smaller the value, the more conducive it is for the country to make stable exports;

[0051] By obtaining the import product information set JK1 and the export product information set CK2 from the customs data of the output-marked country, for product c, its export quantity is CK. 2c The import quantity of the exporting country is JK. 1C Then the complementarity index CK2∩JK1 is the intersection of the set of products exported by the country itself (CK2) and the set of products imported by the country marked by the exporting country (JK1). The sum of the numerators represents the complementary trade volume between the country and the country marked by the exporting country in terms of products, that is, the quantity of products exported and imported between the two countries in terms of products. The larger this value is (the closer it is to 1), the stronger the complementarity between the two countries in terms of products, and the more stable the product exports with the country marked by the exporting country.

[0052] The average correlation coefficient was analyzed using the maximum-minimum normalization method. Normalizing the complementarity index CI yields... and CI G Then the stability score e1 and e2 are the weight coefficients for the corresponding items; a stability score is calculated for output-marked countries initially determined to have stable output, and those with a stability score WD > WD are selected. max The output countries are labeled and sorted according to their scores. WD max The preset stability score threshold is used.

[0053] The competitive data analysis module analyzes the market share, price competitiveness, and product structure matching degree of stable export countries. It also retrieves the correlation coefficients and complementarity indices from the customs data analysis module to comprehensively evaluate its own competitiveness in the market and that of its competitors. The module further filters the stable export countries and intelligently recommends the top-ranked stable export countries based on the ranking of the filtering results.

[0054] For product c, its export quantity is CK. 2c The export volume of competitor m is CKm3c Then, the market share of itself and its competitor D are respectively If FE2>max(FE3), then it is determined that the product c produced by itself has a dominant market share.

[0055] The export transaction amount divided by the export quantity equals the unit price of the exported product. The export unit price of the product itself is DJ2, and the export unit price of the competitor is DJ3. If DJ2≤DJ3 and FE2 shows an upward trend, then the product itself is determined to be more competitive in the market.

[0056] Overlap Index The higher the OI (Overlap Index), the better the match between the product structure and the target market demand, and the stronger the competitiveness.

[0057] The market share percentage (FE2), export unit price (DJ2), and overlap index (OI) were normalized using the maximum-minimum normalization method to obtain FE. 2G DJ 2G and IO G The market competitiveness index ω1, ω2, ω3, ω4, and ω5 are the weighting coefficients for the corresponding items; then, the market competitiveness index CS3 of the competitors is calculated. If CS2 > CS3, it is determined that one's own competitiveness is superior to that of the competitors, and the competitiveness level is [not specified]. The larger the value, the more competitive the company is considered to be compared with its competitors; f is the preset comparison value.

[0058] After obtaining stability scores, the system outputs a ranking of marked countries. Among these marked countries, those with higher competitiveness than their competitors are re-marked, and their competitiveness is calculated. The re-marked countries are then ranked according to their competitiveness and recommended based on their ranking.

[0059] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. An intelligent recommendation method for cross-border markets based on multi-dimensional customs data fusion analysis. Includes, characterized in that, The intelligent recommendation method for cross-border markets is derived from analysis using a cross-border market intelligent recommendation system. The intelligent recommendation method includes the following steps: S1: Classify your own products according to the internationally recognized HS code, establish a "product type hierarchy", and under the "product type hierarchy" establish a "product material hierarchy" by material, and then establish a "product specification hierarchy" by specification parameters. By determining whether the quantity of the parent hierarchy is equal to the total quantity of the child hierarchy, verify the matching of the quantities of each hierarchy to ensure the accuracy of the information. S2: Obtain import volume data for each country within a set time period, calculate and compare the import volume difference between adjacent time points, and mark countries whose import demand is gradually increasing; analyze the historical import volume data of the marked countries, remove outliers, estimate the maximum value of the marked countries, and determine the time required to reach the maximum value based on the estimated maximum value, and determine whether the output to the marked countries is stable. S3: Again, determine whether the relationship with the stable exporting country is stable by using trade relevance and import / export product complementarity index; screen out the countries determined to be stable exporters, and analyze the market share, price competitiveness, and product structure matching degree of the screened countries. The analysis steps are as follows: R1: For product c, its export quantity is CK. 2c The export volume of competitor m is CKm 3c Then, the market share of itself and its competitor D are respectively... , ,like If so, it is determined that the product c produced by itself has a dominant market share. R2: The export transaction amount divided by the export quantity equals the unit price of the exported product. Our own export unit price is DJ2, and the competitor's export unit price is DJ3. If... Furthermore, if FE2 shows an upward trend, it indicates that the product's price is more competitive in the market. R3: Overlap Index The higher the OI (Overlap Index), the better the match between the product structure and the target market demand, and the stronger the competitiveness. R4: After normalization, we get The market competitiveness index , These are the weighting coefficients for the corresponding items; then, the market competitiveness index CS3 of the competitors is calculated, if... If so, it is determined that one's own competitiveness is superior to that of competitors; R5: After obtaining the stability score, output the ranking of the marked countries. Among these marked countries, those with their own competitiveness are better than their competitors. Then, the competitiveness score is calculated and the marked countries are ranked according to the competitiveness score. Recommendations are made according to the ranking.

2. A system for implementing the intelligent recommendation method for cross-border markets based on multi-dimensional customs data fusion analysis as described in claim 1, characterized in that, The intelligent recommendation system includes a customs data analysis module and a competitive data analysis module; The customs data analysis module creates a tree diagram to classify products by type, material, and specifications; it judges the demand of each country based on import volume data, filters countries with high demand, and estimates the maximum import volume of the selected countries by filtering out outliers, and compares it with the import volume growth rate of the selected countries to preliminarily determine whether the country is a stable exporter. The correlation coefficients and complementarity indices of the products with those of various countries are calculated during the trade process, and the initially identified stable export countries are screened again. The competitive data analysis module analyzes the market share, price competitiveness, and product structure matching degree of stable export countries. It also retrieves the correlation coefficients and complementarity indices from the customs data analysis module to comprehensively evaluate its own competitiveness in the market and that of its competitors. The module further filters the stable export countries and intelligently recommends the top-ranked stable export countries based on the ranking of the filtering results. The competitive data analysis module performs the following steps to analyze competitiveness: R1: For product c, its export quantity is CK. 2c The export volume of competitor m is CKm 3c Then, the market share of itself and its competitor D are respectively... , ,like If so, it is determined that the product c produced by itself has a dominant market share. R2: The export transaction amount divided by the export quantity equals the unit price of the exported product. Our own export unit price is DJ2, and the competitor's export unit price is DJ3. If... Furthermore, if FE2 shows an upward trend, it indicates that the product's price is more competitive in the market. R3: Overlap Index The higher the OI (Overlap Index), the better the match between the product structure and the target market demand, and the stronger the competitiveness. R4: After normalization, we get The market competitiveness index , These are the weighting coefficients for the corresponding items; then, the market competitiveness index CS3 of the competitors is calculated, if... If so, it is determined that one's own competitiveness is superior to that of competitors; R5: After obtaining the stability score, output the ranking of the marked countries. Among these marked countries, those with their own competitiveness are better than their competitors. Then, the competitiveness score is calculated and the marked countries are ranked according to the competitiveness score. Recommendations are made according to the ranking.

3. The system according to claim 2, characterized in that, The steps for creating a tree diagram using the customs data analysis module are as follows: K1: In accordance with the internationally accepted Harmonized System for Product Names and Coding, query and record the coding information of its own products, and then establish a corresponding number of tree diagram sub-levels based on the number of its own product coding information, and mark the sub-levels as product type levels; K2: Obtain the material information of the product under the corresponding product code information, and build a corresponding number of tree diagram sub-levels under the product type level based on the number of obtained corresponding product material information, and mark the sub-level as the product material level. K3: Obtain the product specifications under the corresponding product material information, and based on the number of corresponding product specifications obtained, create a corresponding number of tree diagram sub-levels under the product material level, and mark the sub-level as the product specification level. K4: Count the number of items corresponding to each sub-level. If the number of items in the parent level is equal to the total number of items in the sub-levels, the record is considered correct; otherwise, the information of the parent level and all sub-levels under the parent level is recounted.

4. The system according to claim 3, characterized in that, The steps for analyzing the demand for corresponding products using the customs data analysis module are as follows: Q1: Retrieve the coding information of the product with the largest production volume, then obtain the import volume data of the corresponding HS code products in each country within a set time period from the customs data, sort the obtained import volume data of each country according to time, calculate the difference between the import volume data of adjacent time points, and compare the differences. Q2: If the difference gradually increases over time, it is determined that the country's demand for the product is gradually increasing, and the country is marked as an exporter; if the difference gradually decreases over time, it is determined that the country's demand for the product is gradually decreasing; if the fluctuation of the difference over time is within the set fluctuation range, it is determined that the country's demand for the product tends to stabilize.

5. The system according to claim 4, characterized in that, The customs data analysis module's preliminary determination steps for stable export countries are as follows: P1: Obtain historical import data for countries with output markers, calculate the mean A1 and standard deviation B of the obtained historical import data, set the fluctuation range of the import data based on the calculated mean A1 and standard deviation B, and mark the import data outside the set fluctuation range as outliers. P2: After removing outliers, calculate the mean A2 of the remaining import volume data. Use the calculated mean A2 as the historical average import volume data of the corresponding output label country. Calculate the difference between the mean A2 and the extreme value in the historical import volume data, and select the maximum difference as the estimated fluctuation value d. Then the estimated maximum import volume data of that country is A2+d. P3: Obtain the import volume data of the corresponding products of the output marked country within a set time period, sort these import volume data according to time, and substitute the import volume data of the corresponding time data into the formula y=kx+b to obtain the specific values ​​of the coefficients k and b in the formula for the import volume variable data y and the time variable data x. P4: Substituting the maximum value A2+d and the real-time import volume data of the output-marked country for the current time period into y=kx+b, we can deduce that after time x1, the import volume data of the corresponding product of the output-marked country will reach the estimated maximum value. If xs Therefore, it is preliminarily determined that the country with the output mark can stably export the corresponding product. s The preset controllable time is used; otherwise, it is determined that the output of the corresponding product to the country marked with the output label is unstable.

6. The system according to claim 5, characterized in that, The customs data analysis module performs the following secondary determination steps for stable export countries: N1: Let X it To indicate the import volume of similar products by the exporting country during a given period, Y jt Let be the import volume of the j-th major trading country of the world in period t. Then, what is the correlation coefficient between this exporting country and the import volumes of all major trading countries? , To output the average import volume of similar products in country i, Let be the average import volume of similar products in the j-th major trading country, and n be the number of time periods; average correlation coefficient. m represents the number of major trading countries; N2: By obtaining the import product information set JK1 and the export product information set CK2 from the customs data of the output marked country, for product c, the export quantity of the product itself is CK. 2c The import quantity of the exporting country is JK. 1c Then the complementarity index , The intersection of its own export product set CK2 and the export-marked country's import product set JK1; N3: Average correlation coefficient calculated using the maximum-minimum normalization method. Normalizing the complementarity index CI yields... Then the stability score , These are the weight coefficients for the corresponding items; N4: Calculate stability scores for countries initially identified as stable outputs, and select those with stable scores. The output is labeled with the country, and these labeled countries are sorted according to their scores. The preset stability score threshold is used.

Images