Trade process digital optimization method and system

Abstract

The invention relates to the technical field of trade operation, in particular to a trade process digital optimization method and system, and the method comprises the following steps: extracting a time sequence feature value, a logistics node lag time value and an abnormal trade behavior amount of a trade order from a trade data set, and extracting a trade behavior feature range through feature difference operation; according to the invention, through deep analysis of transaction data, extraction of time sequence characteristics of transaction orders and lag time of logistics nodes, accurate screening of transaction behavior characteristics, optimization of data association and improvement of coordination of transaction chains, change conditions of key nodes are dynamically identified based on global data trend analysis, so that the transaction efficiency is improved. According to the method, time sequence coherence is ensured, connectivity between nodes is optimized, through fine-grained time sequence analysis, cooperative capability between the nodes is improved, influence caused by abnormal delay is reduced, a dynamic weight matrix is constructed, potential risks are monitored and identified in real time, path flow and weight distribution are adjusted, optimal configuration of resources is achieved, and pressure of a supply chain is reduced.

Description

Technical Field

[0001] The present invention relates to the field of trade operation technology, and in particular to a trade process digital optimization method and system. Background Art

[0002] The field of trade operation technology includes the management and optimization of trade-related processes, activities and resources. Its core content includes data processing and technical implementation of commodity trading, logistics management, supply chain collaboration, order processing, payment settlement and other links. This technical field improves the efficiency of trade activities through digital means, focusing on establishing an efficient and secure trading system, covering transaction behavior records, commodity and service information management, transaction data statistical analysis, and full-process coordination of the supply chain, aiming to achieve comprehensive digital management and efficient use of trade resources.

[0003] Among them, the digital optimization method of trade processes refers to the improvement and optimization of trade processes through digital technology, covering the classification and organization of transaction information, the digital conversion of trade resources, the automation of transaction processes, and the data docking and processing of trade nodes. This method collects, processes, transmits and integrates information in the trade process to establish an automated system for efficient transactions, and combines specific data sorting methods to achieve real-time monitoring and adjustment of key steps in the transaction process to ensure the consistency and standardization of trade operation processes.

[0004] Existing technologies rely on fixed rules for data processing and cannot flexibly respond to dynamic changes in the transaction process, resulting in weak adaptability of the transaction chain and difficulty in timely adjustment. The recognition accuracy of key transaction nodes is low, and there is a lack of in-depth mining of node timing characteristics, which affects the coordination of the overall process. The adjustment of node connectivity weights is delayed, and the changing trend of transaction data cannot be fully utilized, resulting in unbalanced resource allocation and affecting logistics efficiency. The lack of effective risk transmission monitoring means can easily cause local risks to spread to the entire trade process, increasing business uncertainty. In the process of path optimization, there is a lack of precise adjustment of the dynamic distribution of traffic, which affects the overall smoothness of the transaction chain. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a method and system for digital optimization of trade processes.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: a method for digital optimization of trade processes, comprising the following steps: S1: Extract the time series feature values ​​of transaction orders, the lag time values ​​of logistics nodes and the amount of abnormal transaction behaviors from the trade data set, extract the range of transaction behavior features through feature difference calculation, compare the difference values ​​between data, adjust the feature screening rules to optimize the correlation features between transaction behaviors and logistics nodes, and generate a sparse feature vector set; S2: Based on the sparse feature vector set, calculate the sequence value of the global trade data change trend, extract the dynamic trend of the core transaction nodes, determine whether the lag time of the key nodes meets the standards, analyze the changes in the connectivity weights of the core transaction nodes, verify the consistency, and establish the global key node trend prediction value; S3: Based on the global key node trend prediction value, separate the fine-grained node time change data, calculate the time series correlation measurement value of each node, extract the time change rate value, compare the connectivity weight values ​​of adjacent nodes, call the node update rule to perform correlation iterative optimization, and establish a distributed key node dynamic weight matrix; S4: Based on the distributed key node dynamic weight matrix, extract the risk factor combination value between nodes, calculate the risk transmission probability value, analyze the abnormal change area of ​​the weight matrix, screen the risk-significant nodes, compare the dynamic weight differences between nodes, and establish the node-related risk transmission index; S5: Based on the node-associated risk transfer index, construct an optimized path sequence of risk nodes, adjust the key path flow value and weight value, verify the connectivity of the optimized path, adjust the flow distribution ratio, and generate a key node optimized path set.

[0007] As a further solution of the present invention, the sparse feature vector set includes the transaction behavior feature range, the feature screening rule optimization value, and the transaction behavior and logistics node association characteristics; the global key node trend prediction value includes the dynamic change trend of the core transaction node, the key node lag time standard value, and the core transaction node connectivity weight change; the distributed key node dynamic weight matrix includes fine-grained node time change data, time series association measurement value, time change rate value, and adjacent node connectivity weight value; the node-associated risk transfer index includes the risk factor combination value between nodes, the risk transmission probability value, the weight matrix abnormal change area, the risk-significant node, and the dynamic weight difference between nodes; the key node optimization path set includes the risk node optimization path sequence, the key path flow value, the weight value, and the flow distribution ratio.

[0008] As a further solution of the present invention, the step of acquiring the sparse feature vector set is specifically: S101: extracting time series feature values ​​of transaction orders from the trade data set, and sorting the extracted time series feature values, generating a time difference sequence by calculating the time difference between transaction orders, screening transaction orders in high-frequency transaction time periods according to the distribution characteristics of the time difference sequence, calculating the intersection ratio of the high-frequency transaction time period and the overall transaction time series, and generating a high-frequency transaction feature interval; S102: Cross-comparing the time period feature values ​​in the transaction high-frequency feature interval with the logistics node lag time value, calculating the difference mean and difference standard deviation of each time period according to the interval, selecting intervals whose difference mean is greater than a certain multiple of the standard deviation, and generating a high-difference feature interval set by comparing the interval coverage; S103: Accumulate and calculate the amount of abnormal transaction behaviors in the high-difference feature interval using the formula: ; Calculate the impact values ​​of abnormal trading behaviors in multiple intervals and generate a set of impact values ​​of abnormal trading behaviors; in, Represents the impact value of abnormal trading behavior, Represents the time series characteristic value of the transaction order, Represents the lag time value of the logistics node, Represents the amount of abnormal trading behavior, Represents the total number of high-difference feature intervals; S104: performing sparse processing on the abnormal transaction behavior impact value set, marking the interval where the abnormal transaction behavior impact value is greater than the sparse threshold as a valid interval, and generating a sparse feature vector set by extracting feature parameters in the valid interval.

[0009] As a further solution of the present invention, the step of obtaining the global key node trend prediction value is specifically as follows: S201: extracting and calculating the normalized values ​​of multiple time series in the sparse feature vector set, calculating the dynamic change rate of the time point for the normalized value sequence, screening the prominent change time points based on the dynamic change rate fluctuation value, and generating a global trade data change trend sequence value; S202: Compare the sequence value of the global trade data change trend with the dynamic trend of the core nodes of the trading network, extract the nodes with significant sequence differences, analyze the dynamic trend of the nodes with significant differences in segments, calculate the mean square error value based on the difference between the node rate and the global dynamic trend rate, and generate the dynamic trend of the core trading nodes; S203: Perform standard matching on the lag time of the key nodes of the dynamic trend of the core transaction nodes, and calculate the deviation value of the lag time, using the formula: ; Extract nodes with larger lag time deviation values ​​as lag-exceeding nodes, and generate key nodes with lag time exceeding the standard; in, represents the deviation value of the key node lag time, Represents the lag time of the dynamic trend node, represents the standard time series, Represents the total number of nodes, To adjust the weights, the scaling effect of the lag time standard deviation is Represents the standard deviation of the lag time, which measures the dispersion of the time series. Represents the sum operation of all nodes; S204: Analyze the changes in connectivity weights of the key nodes with excessive lag time and the core transaction nodes, calculate connectivity weight fluctuation values ​​based on the node-by-node weight change rate, screen key nodes associated with fluctuation values ​​and lag effects, and establish global key node trend prediction values.

[0010] As a further solution of the present invention, the steps of obtaining the distributed key node dynamic weight matrix are specifically as follows: S301: Based on the global key node trend prediction value, refine the time change data of each node, extract the time fine-grained changes of multiple nodes through time series decomposition, analyze the change trend in combination with the time point information, and generate time fine-grained node change data; S302: performing time series correlation measurement on the time fine-grained node change data, calculating the time correlation coefficient between adjacent nodes based on the time series of the nodes, selecting nodes with strong correlation through correlation analysis, and generating a time series correlation measurement value for each node; S303: extracting the time change rate in the time series association metric value, calculating the change rate of the node time series by using a differential algorithm, screening and marking nodes with large rate differences, and generating a time change rate value; S304: Compare the time change rate value with the connectivity weight value of the adjacent nodes using the formula: ; Call the node update rules, adjust the weights node by node, and establish a distributed dynamic weight matrix of key nodes; in, represents the updated node weight, represents the original node weight, is the adjustment coefficient, which is used to control the impact of rate changes on weight adjustment. represents the rate difference between nodes, is the average node rate.

[0011] As a further solution of the present invention, the step of obtaining the node-associated risk transfer indicator is specifically as follows: S401: extracting the risk factor combination value between nodes based on the distributed key node dynamic weight matrix, analyzing the dynamic change rate of multiple nodes in the weight matrix, and calculating the risk correlation between multiple nodes in combination with the risk factor impact value to generate the risk factor combination value between nodes; S402: Calculate the transmission probability of the risk factor combination value between the nodes, select the node group according to the risk correlation, analyze the transmission path and probability distribution of the risk factor between the nodes, calculate the risk transmission probability in combination with the cumulative transmission model, and generate the risk transmission probability value between the nodes; S403: Compare and analyze the risk transmission probability value between the nodes and the abnormal change area of ​​the weight matrix, and calculate the correlation between the risk transmission probability and the weight change, using the formula: ; The risk transmission probability is weighted based on the weight change rate to screen out risk-significant nodes; in, Represents the association value of the risk-significant node, For the The transmission probability of a node is For the The weight value of the node, is the change in node weight, is the average weight of the nodes, To adjust the impact of changes in weights of factor controls; S404: Analyze the dynamic weight differences of the risk-significant nodes, compare the matching degree between the dynamic weight changes and the transmission probability values ​​between the nodes, calculate the risk transfer relationship between the nodes in combination with the risk correlation, and establish a node-related risk transfer index.

[0012] As a further solution of the present invention, the step of obtaining the key node optimization path set is specifically: S501: Based on the node-associated risk transmission index, by analyzing the risk transmission probability and weight change value between nodes one by one, the key nodes with higher transmission probability among the risk nodes are screened, and the initial optimization path is determined in combination with the path flow characteristics to generate an optimized path sequence of the risk nodes; S502: adjusting the key path flow value and weight value of the optimized path sequence of the risk node, calculating the contribution weights of multiple nodes based on the node flow ratio in the path, redistributing the flow value by the total weight difference rate, calculating the adjusted path connectivity index, and adjusting the key path flow value and weight value; S503: Verify the optimized path connectivity after adjusting the critical path flow value and weight value, by calculating the impact of the flow distribution ratio on the path connectivity, using the formula: ; Standardize and optimize the distribution of node traffic on the path, generate optimized traffic distribution based on the weight adjustment ratio, and adjust the traffic distribution ratio; in, Represents the path connectivity index, which measures the impact of traffic distribution and weight between nodes in the path. For the path The traffic value of a node indicates the transmission capacity of the node. is the weight value of the corresponding node, indicating the importance of the node. The total number of nodes on the path, providing a reference for the path size, is an adjustment factor used to control the impact of traffic fluctuations on path connectivity. It represents the standard deviation of the path node traffic and measures the uniformity of node traffic distribution; S504: Optimizing feature analysis is performed on the path after adjusting the traffic distribution ratio, by calculating the traffic balance and node connectivity matching rate within the path, selecting the path according to traffic distribution balance and connectivity, and generating a key node optimized path set according to the optimization rules.

[0013] A trade process digital optimization system, the trade process digital optimization system is used to execute the trade process digital optimization method, the system comprises: The data feature extraction module extracts the time series feature values ​​of transaction orders, the lag time values ​​of logistics nodes and the amount of abnormal transaction behaviors based on the trade data set, calculates the differences of multiple feature values, and calculates the range of transaction behavior features, compares and adjusts the feature screening rules, optimizes the correlation features between transaction behaviors and logistics nodes, and generates a sparse feature vector set; The global trend analysis module calculates the global trade data change trend sequence value based on the sparse feature vector set, extracts the dynamic change trend of the core transaction nodes, determines whether the key node lag time value meets the standard, analyzes the change of the core transaction node connectivity weight, verifies the consistency, and establishes the global key node trend prediction value; The key node optimization module separates the fine-grained node time change data based on the global key node trend prediction value, calculates the time series correlation measurement value of each node, extracts the time change rate value, compares the connectivity weight values ​​of adjacent nodes, calls the node update rule to iteratively optimize the node correlation, and establishes a distributed key node dynamic weight matrix; The risk factor analysis module extracts the risk factor combination value between nodes based on the distributed key node dynamic weight matrix, calculates the risk transmission probability value, analyzes the abnormal change area in the weight matrix, screens the risk-significant nodes, compares the dynamic weight differences between nodes, and establishes the node-related risk transmission index; The path optimization module constructs an optimized path sequence of risk nodes based on the node-associated risk transfer index, adjusts the key path flow value and weight value, verifies the connectivity of the optimized path, adjusts the flow distribution ratio, and generates a key node optimized path set.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by deeply analyzing transaction data, extracting the timing characteristics of transaction orders and the lag time of logistics nodes, accurately screening transaction behavior characteristics, optimizing data association, and improving the coordination of the transaction chain. Based on global data trend analysis, dynamically identify changes in key nodes, ensure timing continuity and optimize connectivity between nodes. Through fine-grained timing analysis, improve the coordination ability between nodes and reduce the impact of abnormal delays. Construct a dynamic weight matrix, monitor and identify potential risks in real time, adjust path flow and weight distribution, achieve optimal resource allocation, reduce supply chain pressure, and improve overall operational efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of the workflow of the present invention; Figure 2 A flow chart of the steps for obtaining a sparse feature vector set of the present invention; Figure 3 A flowchart of steps for obtaining the global key node trend prediction value of the present invention; Figure 4 A flowchart of the steps for obtaining the dynamic weight matrix of distributed key nodes of the present invention; Figure 5 A flowchart of the steps for obtaining the node-associated risk transfer indicator of the present invention; Figure 6 A flow chart of the steps for obtaining a set of key node optimization paths of the present invention. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0017] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined. Embodiment 1

[0018] See also Figure 1 The present invention provides a technical solution: a method for digital optimization of trade processes, comprising the following steps: S1: Extract the time series feature values ​​of transaction orders, the lag time values ​​of logistics nodes and the amount of abnormal transaction behaviors from the trade data set, extract the range of transaction behavior features through feature difference calculation, compare the difference values ​​between data, adjust the feature screening rules to optimize the correlation features between transaction behaviors and logistics nodes, and generate a sparse feature vector set; S2: Based on the sparse feature vector set, calculate the sequence value of the global trade data change trend, extract the dynamic trend of the core transaction nodes, determine whether the lag time of the key nodes meets the standards, analyze the changes in the connectivity weights of the core transaction nodes, verify the consistency, and establish the global key node trend prediction value; S3: Based on the global key node trend prediction value, separate the fine-grained node time change data, calculate the time series correlation measurement value of each node, extract the time change rate value, compare the connectivity weight values ​​of adjacent nodes, call the node update rule to perform correlation iterative optimization, and establish a distributed key node dynamic weight matrix; S4: Based on the dynamic weight matrix of distributed key nodes, extract the combined value of risk factors between nodes, calculate the risk transmission probability value, analyze the abnormal change area of ​​the weight matrix, screen the risk-significant nodes, compare the dynamic weight differences between nodes, and establish the node-related risk transmission index; S5: Based on the node-associated risk transfer indicators, construct the optimized path sequence of risk nodes, adjust the key path flow value and weight value, verify the connectivity of the optimized path, adjust the flow distribution ratio, and generate a set of key node optimized paths.

[0019] The sparse feature vector set includes the transaction behavior feature range, feature screening rule optimization value, and transaction behavior and logistics node association features. The global key node trend prediction value includes the dynamic change trend of core transaction nodes, the key node lag time standard value, and the core transaction node connectivity weight change. The distributed key node dynamic weight matrix includes fine-grained node time change data, time series association measurement value, time change rate value, and adjacent node connectivity weight value. The node-related risk transfer indicators include the risk factor combination value between nodes, the risk transmission probability value, the weight matrix abnormal change area, the risk-significant node, and the dynamic weight difference between nodes. The key node optimization path set includes the risk node optimization path sequence, the key path flow value, the weight value, and the flow distribution ratio.

[0020] See also Figure 2 , the specific steps for obtaining the sparse feature vector set are: S101: extracting time series feature values ​​of transaction orders from the trade data set, and sorting the extracted time series feature values, generating a time difference sequence by calculating the time difference between transaction orders, screening transaction orders in high-frequency transaction time periods according to the distribution characteristics of the time difference sequence, calculating the intersection ratio of the high-frequency transaction time period and the overall transaction time series, and generating a high-frequency transaction feature interval; Extract the time series feature values ​​of transaction orders from the trade data set. First, classify the order data in chronological order, generate a time difference sequence by calculating the time difference between each transaction order, and use the time difference sequence as the basic feature quantity to measure the transaction time interval. Then analyze the distribution characteristics of the time difference sequence, and use the cumulative distribution function (CDF) to map the time difference sequence to different time intervals, so as to determine the distribution range of the high-frequency trading time period. Then, calculate the intersection ratio of the number of transaction orders in the high-frequency trading time period and the total number of transaction orders. The calculation method is the ratio formula: ,in is the number of orders in the high-frequency time period, is the total number of transaction orders. Finally, the calculated intersection ratio result is combined with the time period feature value to obtain and mark the range characteristics of the high-frequency trading time period and generate the high-frequency feature interval of the transaction.

[0021] S102: Cross-comparing the time period feature values ​​in the transaction high-frequency feature interval with the logistics node lag time value, calculating the difference mean and difference standard deviation of each time period according to the interval, selecting the intervals whose difference mean is greater than a certain multiple of the standard deviation, and generating a high-difference feature interval set by comparing the interval coverage; The time period characteristic values ​​in the high-frequency transaction feature interval are compared with the logistics node lag time values. The absolute difference between the time characteristic values ​​of the high-frequency transaction time period and the logistics node lag time is calculated one by one. The absolute difference calculation results of each time period are averaged and the standard deviation is calculated. The formulas are: And the standard deviation formula: , screen out the time periods where the mean of the absolute difference is greater than a certain multiple of the standard deviation to form high-difference feature intervals, and generate a high-difference feature interval set by superimposing the coverage of the high-difference intervals and comparing the comparison features of the subsequent transaction node information.

[0022] S103: Accumulate and calculate the amount of abnormal transaction behaviors concentrated in the high-difference feature interval using the formula: ; Calculate the impact values ​​of abnormal trading behaviors in multiple intervals and generate a set of impact values ​​of abnormal trading behaviors; in, Represents the impact value of abnormal trading behavior, Represents the time series characteristic value of the transaction order, Represents the lag time value of the logistics node, Represents the amount of abnormal trading behavior, Represents the total number of high-difference feature intervals; formula: ; The benefit of the formula is that by combining the timing characteristic values ​​of transaction orders, the lag time values ​​of logistics nodes and the amount of abnormal transaction behavior, it can accurately quantify the impact value of abnormal transaction behavior in each time period, providing accurate data support for subsequent optimization.

[0023] Detailed explanation of the formula and the process of formula calculation and derivation: Represents the time series characteristic value of the transaction order, which is obtained by monitoring the daily transaction time point records; It represents the lag time value of the logistics node, which is calculated by extracting the difference between the actual processing time and the planned time of the corresponding node; The amount of abnormal trading behavior requires screening and accumulation of trading behavior data, which is obtained by calculating the total number of trading behaviors that do not meet expectations every day; is the number of high-difference feature intervals, which is obtained by counting the high-difference feature interval set generated in step 2.

[0024] Assignment: Assumption is [10,12,14] hours, is [8,11,13] hours, is [3,5,4], , formula derivation: ; The result shows that the impact value of abnormal trading behavior in a time period is 5. By analyzing the cumulative impact of each trading interval, it can provide a reference for the next step of sparse processing.

[0025] S104: Performing sparse processing on the abnormal transaction behavior impact value set, marking the interval where the abnormal transaction behavior impact value is greater than the sparse threshold as a valid interval, and generating a sparse feature vector set by extracting feature parameters in the valid interval.

[0026] The set of abnormal trading behavior impact values ​​is sparsely processed, and the abnormal trading behavior impact values ​​are screened according to a certain sparse threshold. First, the distribution density of the abnormal trading behavior impact value set is calculated, and the value range is divided by the distribution density function. The interval with an impact value greater than a specific sparse threshold is extracted as the valid interval, and then the characteristic parameters of the trading orders in each valid interval are extracted and analyzed. Finally, a sparse feature vector set is constructed by minimizing the parameter dimension.

[0027] See also Figure 3 ,The specific steps for obtaining the trend prediction value of the global key node are: S201: extracting and calculating the normalized values ​​of multiple time series in the sparse feature vector set, calculating the dynamic change rate of the time point for the normalized value sequence, screening the prominent change time points based on the dynamic change rate fluctuation value, and generating the global trade data change trend sequence value; The global trade data change trend sequence value is extracted from the sparse feature vector set, and the normalized value of each time series in the sparse feature vector is calculated. The normalized value is obtained by normalizing each feature of the time point sequence point by point to ensure that the change rate of the feature is calculated within the same dimension. The dynamic change rate calculation needs to use the value difference of consecutive time points in the time series as the numerator and the normalized time point value as the denominator to obtain the change rate value. The change rate fluctuation value screening needs to be classified according to the difference amount, and a fluctuation threshold based on the normalized range is set. The significant fluctuation point screening is performed by marking and extracting the points with fluctuation values ​​greater than the threshold. After completing the above steps, the marked significant fluctuation points are summarized as a significant change time point set to generate the global trade data change trend sequence value.

[0028] S202: Compare the sequence value of the trend of global trade data changes with the dynamic trend of the core nodes of the trading network, extract the nodes with significant sequence differences, analyze the dynamic trend of the nodes with significant differences in segments, calculate the mean square error value based on the difference between the node rate and the global dynamic trend rate, and generate the dynamic trend of the core trading nodes; The sequence values ​​of the trend of changes in global trade data are compared with the dynamic trends of the core nodes of the trading network. First, the dynamic trends of the core nodes need to be realized by extracting the time series characteristics of the key nodes in the trading network. The extraction of dynamic trends requires segmenting the time series and calculating the change rate of each segment. The rate value is obtained by dividing the difference in the characteristic values ​​of the two end time points by the time span. When extracting nodes with significant sequence differences, it is necessary to calculate the mean square error of the change rate of each node and the global dynamic trend rate. The screening of significant nodes is achieved by setting the difference threshold of the mean square error. The segmentation analysis requires detailed segmentation of the significant node time series, and analysis combined with the rate change characteristics of each segment to generate the dynamic trend of the core trading nodes.

[0029] S203: Standard matching is performed on the lag time of the key nodes of the dynamic trend of the core transaction nodes, and the deviation value of the lag time is calculated using the formula: ; Extract nodes with larger lag time deviation values ​​as lag-exceeding nodes, and generate key nodes with lag time exceeding the standard; in, represents the deviation value of the key node lag time, Represents the lag time of the dynamic trend node, represents the standard time series, Represents the total number of nodes, To adjust the weights, the scaling effect of the lag time standard deviation is Represents the standard deviation of the lag time, which measures the dispersion of the time series. Represents the sum operation of all nodes; formula: ; The usefulness of the formula is that by adding the adjustment weight and standard deviation The combined effect of the two improves the sensitivity of the deviation value to the distribution characteristics and outliers, thereby more accurately evaluating the deviation of the key node lag time.

[0030] Detailed explanation of the formula and the process of formula calculation and derivation: Assume the lag time to be , standard time is , weight , calculate the standard deviation , the sum is , the deviation value is calculated as: ; The result shows that the deviation value of the lag time is 0.37, which reflects the degree of lag time deviation of the key nodes. It can be further used to screen the key nodes with large lag time deviation as analysis objects and generate the key nodes with excessive lag time.

[0031] S204: Analyze the changes in connectivity weights of key nodes and core transaction nodes with excessive lag time, calculate connectivity weight fluctuation values ​​based on the node-by-node weight change rate, screen key nodes associated with fluctuation values ​​and lag effects, and establish global key node trend prediction values.

[0032] A comprehensive analysis is conducted on the changes in connectivity weights of key nodes with excessive lag time and core transaction nodes. The correlation between the lag time value and the weight change rate of the node with excessive lag time is calculated. The acquisition of connectivity weight changes requires recording the transaction volume characteristics of each node and normalizing the transaction volume change sequence. The normalized change sequence is used to calculate the weight change rate of the node. The rate value is calculated by dividing the change between the two points before and after the sequence by the time point interval. The weight change fluctuation value is calculated based on the correlation fitting function between the rate and the lag time. When screening nodes with significant fluctuation values, the screening threshold is set based on the fitting residual value to establish a global key node trend prediction value.

[0033] See also Figure 4 ,The specific steps for obtaining the dynamic weight matrix of distributed key nodes are: S301: Based on the global key node trend prediction value, refine the time change data of each node, extract the time fine-grained changes of multiple nodes through time series decomposition, analyze the change trend in combination with the time point information, and generate time fine-grained node change data; Based on the global key node trend prediction value, extract the time change data of each node, and decompose the time series data into fine-grained time change points by node by constructing the time series of the node. Combined with the attribute data of the node, further subdivide the characteristic value of each time point in the time series. According to the change rate between different time points, compare the growth or decay trend of the time series between nodes, use the change amplitude value to segment and classify the time series, and generate the change trend curve of fine-grained time points through segment accumulation. Combined with the time point correlation relationship between nodes, integrate the data by time segment to obtain the change data of the node time segment. Based on the data, filter out the node time data with fluctuation characteristics, extract the unique trend of key nodes in time change, and generate time fine-grained node change data; S302: Perform time series correlation measurement on the time fine-grained node change data, calculate the time correlation coefficient between adjacent nodes based on the time series of the nodes, select nodes with strong correlation through correlation analysis, and generate a time series correlation measurement value for each node; The time series correlation measurement is performed on the time-granular node change data. Based on the time point distribution between adjacent nodes of the time series, the correlation matrix is ​​constructed. By calculating the correlation coefficient of the time series between nodes, the node pairs with significant correlation are screened, and the significantly correlated node relationships in the correlation matrix are marked as strongly correlated. The difference value of the time series is calculated in combination with the time point trend curve of the node. The time correlation between the nodes is judged by setting a threshold for the difference value. The time correlation model is further established based on the time distribution characteristics of the nodes. The node connectivity relationship in the model is adjusted. The time correlation coefficient of the node is updated according to the distribution of the correlation matrix. Finally, the time point data and correlation coefficient of each node are integrated to generate the time series correlation measurement value of each node. S303: extracting the time change rate in the time series correlation metric value, calculating the change rate of the node time series by using a differential algorithm, screening and marking nodes with large rate differences, and generating a time change rate value; Extract the time change rate from the time series correlation metric value, calculate the time series change rate of each node through the difference algorithm, compare the rate differences of each node according to the volatility of the node time change rate, filter out the nodes with large rate differences based on the rate fluctuation value, set the rate change threshold, mark the nodes whose rate change exceeds the threshold as rate abnormal nodes, classify the time series fluctuation characteristics of the rate abnormal nodes, generate a set of nodes with significant rate differences through classification statistics, calculate the slope value of the rate change curve, analyze the change trend of the time rate of each node based on the slope value, integrate the rate change trend of the abnormal nodes, and generate the time change rate value; S304: Compare the time change rate value with the connectivity weight value of the adjacent node using the formula: ; Call the node update rules, adjust the weights node by node, and establish a distributed dynamic weight matrix of key nodes; in, represents the updated node weight, represents the original node weight, is the adjustment coefficient, which is used to control the impact of rate changes on weight adjustment. represents the rate difference between nodes, is the average node rate.

[0034] formula: ; The benefit of the formula is that by introducing the rate change With average rate ratio, combined with the adjustment factor , effectively reflects the dynamic impact of rate changes between adjacent nodes on weight adjustment, and improves the sensitivity and accuracy of weight updates; Detailed explanation of the formula and the process of formula calculation and derivation: Assume that the initial node weight is , adjustment coefficient , node rate change , the average node rate , put it into the formula: ; Calculate the internal ratio: ; Calculate the adjustment: ; Calculate the updated weight value: ; This result shows that the node weights have increased from the initial value Updated to ,The updated weight values ​​comprehensively consider the rate changes and the ,dynamic characteristics of nodes, which helps to construct the dynamic weight matrix of ,distributed key nodes.

[0035] See also Figure 5 ,The specific steps for obtaining the node-associated risk transfer index are: S401: extracting the risk factor combination value between nodes based on the dynamic weight matrix of distributed key nodes, calculating the risk correlation between multiple nodes in the weight matrix by analyzing the dynamic change rate of multiple nodes in combination with the risk factor impact value, and generating the risk factor combination value between nodes; Based on the dynamic weight matrix of distributed key nodes, the combined value of risk factors between nodes is extracted, the dynamic change rate of each node in the weight matrix is ​​analyzed, and the relative amplitude and cumulative value of each change in the time series are calculated node by node. The formula is used: Quantify the changes in each node, Indicates The weight change value of each node in the weight matrix, It is the average value of all node weight values. According to the calculation result of weight change rate, the node range with change rate higher than the threshold is screened out. Combined with the impact range data of risk factors, the risk factor combination is matched item by item for nodes with high change rate, and the risk factors are quantified as impact probability values. For example, the economic volatility factor takes the range of 0.3 to 0.5, and the market delay factor takes the range of 0.1 to 0.2. The risk correlation is matched according to the dynamic change ratio, and the risk correlation combination value between the two nodes is calculated. The average value of the dynamic change rate is multiplied by the average value of the risk factor impact range, and the combination correlation weight is calculated by adjusting the multiple. The weight value is stored as the quantified result of the risk factor combination to generate the risk factor combination value between nodes.

[0036] Indicates The weight change rate of each node is used to quantify the relative degree of weight change of the node in the time series; Indicates The weight change of each node is calculated by the difference of the node weight matrix at different times; Represents the average value of all node weights, which is used to standardize the impact of weight changes. The calculation formula is: ; in, For the The weight value of the node, is the total number of nodes; The risk factor impact value is the quantitative result of specific risk variables (such as economic fluctuations, market delays, etc.), which is determined by analyzing the historical data range and impact intensity.

[0037] Assume that the node weight is , , , the node weight change is , , , calculate the weighted average: ; Substituting into the formula: ; Calculated: ; Match the portfolio weights according to the impact range of risk factors. For example, the economic volatility factor is 0.4, the market delay factor is 0.15, and the portfolio correlation weights are calculated for nodes with high change rates: ; Calculated: ; The result shows that the correlation between the node change rate and the risk factor is quantified as 0.09005, which can be further used for risk transfer analysis between nodes.

[0038] S402: Calculate the transmission probability of the risk factor combination value between nodes, select the node group according to the risk correlation, analyze the transmission path and probability distribution of the risk factor between nodes, calculate the risk transmission probability in combination with the cumulative transmission model, and generate the risk transmission probability value between nodes; Calculate the transmission probability of the risk factor combination value between nodes, identify the node group with greater risk correlation, analyze the transmission path and probability distribution pair by pair, and use the quantitative risk factor combination value as an input parameter when calculating the transmission probability value of each node. Calculate, where is the risk factor impact weight, is the node risk change rate. By monitoring the sum of the risks of each node in the conduction path, the key nodes in the conduction path are screened, and the conduction probability of each node is determined according to the distribution interval of the conduction probability value. For example, the conduction probability of multiple nodes is calculated. , , , mark the nodes below the average value as nodes with weaker conduction, extract path combinations for nodes with high conduction probability, calculate the comprehensive conduction value of key nodes using the sum of conduction paths, and generate the risk conduction probability value between nodes.

[0039] S403: Compare and analyze the risk transmission probability value between nodes and the abnormal change area of ​​the weight matrix, and calculate the correlation between the risk transmission probability and the weight change, using the formula: ; The risk transmission probability is weighted based on the weight change rate to screen out risk-significant nodes; in, Represents the association value of the risk-significant node, For the The transmission probability of a node is For the The weight value of the node, is the change in node weight, is the average weight of the nodes, To adjust the impact of changes in weights of factor controls; formula: ; The benefit of the formula is that by dynamically adjusting the transmission probability, weight change rate and the combined effect of the adjustment factor, it can accurately capture the changing characteristics of risk-significant nodes and enhance the system's ability to accurately quantify risk transmission.

[0040] Detailed explanation of the formula and the process of formula calculation and derivation: By monitoring the risk transmission path, Through the weight matrix collection, By calculating the weight change value, is the average node weight, To adjust the factor, substitute into the formula: ; Calculating item by item: ; ; The results show that the comprehensive correlation of risk-significant nodes is 1.270, which is out of range compared with the benchmark value of 0.8, indicating that these nodes have higher risks in weight changes and conduction paths, and are further used to screen risk-significant nodes.

[0041] S404: Analyze the dynamic weight differences of risk-significant nodes, compare the matching degree between the dynamic weight changes and the transmission probability values ​​between nodes, calculate the risk transfer relationship between nodes in combination with the risk correlation, and establish a node-related risk transfer index.

[0042] The dynamic weight differences of risk-significant nodes are analyzed. The dynamic weight changes of nodes are compared pair by pair. The matching relationship between nodes is evaluated in combination with the transmission probability value. For nodes with significant dynamic weight changes, nodes with higher change rates are used as initial input parameters, and the transmission paths between nodes are accumulated. The weight differences in the transmission paths are comprehensively considered, and the risk transmission correlation of each node is calculated through the cumulative value of the dynamic transmission path. The node correlation is mapped to a specific transmission index value, and the transmission order of nodes is adjusted pair by pair to optimize the final index, thus establishing a node-associated risk transmission index.

[0043] See also Figure 6 , the specific steps for obtaining the key node optimization path set are: S501: Based on the node-related risk transmission index, the risk transmission probability and weight change value between nodes are analyzed node by node, and the key nodes with higher transmission probability among the risk nodes are screened. The initial optimization path is determined in combination with the path flow characteristics, and the optimization path sequence of the risk nodes is generated; Based on the node-associated risk transmission indicators, the risk transmission probability and weight change value between nodes are analyzed node by node. First, the transmission probability value of each node on the path is collected, and the data flow value transmitted between nodes is recorded by the monitoring system. The risk transmission probability is ranked according to the dynamic weight change rate, and the top 20% of the nodes are selected as the key node group. Then, the flow fluctuation amplitude and node connectivity are analyzed in the key node group, and the nodes with poor flow stability are screened out through the standard deviation calculation formula of the fluctuation amplitude. The nodes with poor stability are given priority to be included in the initial optimization path. Finally, the flow proportion between nodes is adjusted according to the balance standard of the path flow, and the initial optimization path is determined according to the standardized results of the flow of all nodes on the path to generate the optimized path sequence of risk nodes.

[0044] S502: adjusting the key path flow value and weight value of the optimized path sequence of the risk node, calculating the contribution weights of multiple nodes based on the node flow ratio in the path, redistributing the flow value by the total weight difference rate, calculating the adjusted path connectivity index, and adjusting the key path flow value and weight value; The optimized path sequence of risk nodes is adjusted for critical path flow values ​​and weight values, and the contribution weight of each node is calculated based on the node flow ratio in the path. The node flow values ​​in the optimized path are first counted, and the relative flow ratio of each node in the path is calculated using the standardized processing formula. The contribution weight is calculated by multiplying the ratio by the initial weight value of the node. Then, the difference rate between the sum of the node weights in the path and the target weight is compared, and the flow values ​​of the nodes in the path are redistributed according to the weight difference ratio. At the same time, the connectivity calculation formula is used to verify the overall connectivity level of the path after adjustment. The adjusted flow value is compared with the initial value to confirm the rationality of the allocation, and the critical path flow value and weight value are adjusted.

[0045] S503: Verify the optimized path connectivity after adjusting the critical path flow value and weight value, by calculating the impact of the flow distribution ratio on the path connectivity, using the formula: ; Standardize and optimize the distribution of node traffic on the path, generate optimized traffic distribution based on the weight adjustment ratio, and adjust the traffic distribution ratio; in, Represents the path connectivity index, which measures the impact of traffic distribution and weight between nodes in the path. For the path The traffic value of a node indicates the transmission capacity of the node. is the weight value of the corresponding node, indicating the importance of the node. The total number of nodes on the path, providing a reference for the path size, is an adjustment factor used to control the impact of traffic fluctuations on path connectivity. It represents the standard deviation of the path node traffic and measures the uniformity of node traffic distribution; formula: ; The benefit of the formula is that by adding the node flow standard deviation term Reflect the uniformity of flow distribution and introduce The adjustment factor dynamically adjusts the weight of traffic fluctuations on the connectivity index, improving the sensitivity of path connectivity optimization.

[0046] Detailed explanation of the formula and the process of formula calculation and derivation: ; It is assumed that the flow values ​​of the five nodes on the path are , , , , The node weights are , , , , , flow standard deviation The calculated value is 25, the adjustment factor , the calculation process is as follows: 1. Find the sum of the product of weight and flow: ; 2. Find the flow and: ; 3. Calculate the adjustment: ; 4. Calculate connectivity indicators: ; The result shows that the path connectivity index is 0.791, which is close to 1, indicating that the path flow and weight distribution are relatively balanced. The balance of flow distribution can be further improved through optimization, and finally the stable connectivity optimization of the path can be achieved.

[0047] S504: Optimizing the path after adjusting the traffic distribution ratio, analyzing the optimization characteristics, calculating the traffic balance and node connectivity matching rate within the path, selecting the path based on traffic distribution balance and connectivity, and generating a key node optimized path set according to the optimization rules.

[0048] The optimization characteristics of the paths after adjusting the traffic distribution ratio are analyzed. By calculating the traffic balance degree and node connectivity matching rate within the path, the balance is calculated using the traffic variance formula for the adjusted node traffic distribution. A smaller variance indicates a balanced distribution. The paths with a traffic variance within 5% are taken as alternatives. Then, the connectivity matching rate between nodes is calculated. A matching rate greater than 90% indicates that the path stability is high. Paths with higher matching rates are screened to generate an optimized path set. Finally, the paths that meet the conditions of balance and connectivity are cross-analyzed, and the intersection part is selected as the final result to generate an optimized path set for key nodes.

[0049] A trade process digital optimization system, which is used to execute the trade process digital optimization method, comprises: The data feature extraction module extracts the time series feature values ​​of transaction orders, the lag time values ​​of logistics nodes and the amount of abnormal transaction behaviors based on the trade data set, calculates the differences of multiple feature values, and calculates the range of transaction behavior features, compares and adjusts the feature screening rules, optimizes the correlation features between transaction behaviors and logistics nodes, and generates a sparse feature vector set; The global trend analysis module calculates the trend sequence value of global trade data based on the sparse feature vector set, extracts the dynamic trend of core transaction nodes, determines whether the lag time value of key nodes meets the standard, analyzes the change of connectivity weight of core transaction nodes, verifies the consistency, and establishes the global key node trend prediction value; The key node optimization module is based on the global key node trend prediction value, separates the fine-grained node time change data, calculates the time series correlation measurement value of each node, extracts the time change rate value, compares the connectivity weight values ​​of adjacent nodes, calls the node update rule to iteratively optimize the node correlation, and establishes a distributed key node dynamic weight matrix; The risk factor analysis module is based on the dynamic weight matrix of distributed key nodes, extracts the risk factor combination value between nodes, calculates the risk transmission probability value, analyzes the abnormal change area in the weight matrix, screens the risk-significant nodes, compares the dynamic weight differences between nodes, and establishes the node-related risk transmission index; The path optimization module constructs an optimized path sequence of risk nodes based on the node-associated risk transfer indicators, adjusts the key path flow value and weight value, verifies the connectivity of the optimized path, adjusts the flow distribution ratio, and generates a set of optimized paths for key nodes.

[0050] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for digital optimization of trade processes, characterized in that: The following steps are involved: S1: Extract the time series feature values ​​of transaction orders, the lag time values ​​of logistics nodes and the amount of abnormal transaction behaviors from the trade data set, extract the range of transaction behavior features through feature difference calculation, compare the difference values ​​between data, adjust the feature screening rules to optimize the correlation features between transaction behaviors and logistics nodes, and generate a sparse feature vector set; S2: Based on the sparse feature vector set, calculate the sequence value of the global trade data change trend, extract the dynamic trend of the core transaction nodes, determine whether the lag time of the key nodes meets the standards, analyze the changes in the connectivity weights of the core transaction nodes, verify the consistency, and establish the global key node trend prediction value; S3: Based on the global key node trend prediction value, separate the fine-grained node time change data, calculate the time series correlation measurement value of each node, extract the time change rate value, compare the connectivity weight values ​​of adjacent nodes, call the node update rule to perform correlation iterative optimization, and establish a distributed key node dynamic weight matrix; S4: Based on the distributed key node dynamic weight matrix, extract the risk factor combination value between nodes, calculate the risk transmission probability value, analyze the abnormal change area of ​​the weight matrix, screen the risk-significant nodes, compare the dynamic weight differences between nodes, and establish the node-related risk transmission index; S5: Based on the node-associated risk transfer index, construct an optimized path sequence of risk nodes, adjust the key path flow value and weight value, verify the connectivity of the optimized path, adjust the flow distribution ratio, and generate a key node optimized path set.

2. The method for digital optimization of trade processes according to claim 1, characterized in that: The sparse feature vector set includes the transaction behavior feature range, feature screening rule optimization value, and transaction behavior and logistics node association features. The global key node trend prediction value includes the dynamic change trend of core transaction nodes, the key node lag time standard value, and the core transaction node connectivity weight change. The distributed key node dynamic weight matrix includes fine-grained node time change data, time series association measurement value, time change rate value, and adjacent node connectivity weight value. The node-associated risk transfer index includes the risk factor combination value between nodes, the risk transmission probability value, the weight matrix abnormal change area, the risk-significant node, and the dynamic weight difference between nodes. The key node optimization path set includes the risk node optimization path sequence, the key path flow value, the weight value, and the flow distribution ratio.

3. The method for digital optimization of trade processes according to claim 2, characterized in that: The steps of obtaining the sparse feature vector set are specifically as follows: S101: extracting time series feature values ​​of transaction orders from the trade data set, and sorting the extracted time series feature values, generating a time difference sequence by calculating the time difference between transaction orders, screening transaction orders in high-frequency transaction time periods according to the distribution characteristics of the time difference sequence, calculating the intersection ratio of the high-frequency transaction time period and the overall transaction time series, and generating a high-frequency transaction feature interval; S102: Cross-comparing the time period feature values ​​in the transaction high-frequency feature interval with the logistics node lag time value, calculating the difference mean and difference standard deviation of each time period according to the interval, selecting intervals whose difference mean is greater than a certain multiple of the standard deviation, and generating a high-difference feature interval set by comparing the interval coverage; S103: Accumulate and calculate the amount of abnormal transaction behaviors in the high-difference feature interval, using the formula: ； Calculate the impact values ​​of abnormal trading behaviors in multiple intervals and generate a set of impact values ​​of abnormal trading behaviors; in, Represents the impact value of abnormal trading behavior, Represents the time series characteristic value of the transaction order, Represents the lag time value of the logistics node, Represents the amount of abnormal trading behavior, Represents the total number of high-difference feature intervals; S104: performing sparse processing on the abnormal transaction behavior impact value set, marking the interval where the abnormal transaction behavior impact value is greater than the sparse threshold as a valid interval, and generating a sparse feature vector set by extracting feature parameters in the valid interval.

4. The method for digital optimization of trade processes according to claim 3, characterized in that: The steps for obtaining the global key node trend prediction value are specifically as follows: S201: extracting and calculating the normalized values ​​of multiple time series in the sparse feature vector set, calculating the dynamic change rate of the time point for the normalized value sequence, screening the prominent change time points based on the dynamic change rate fluctuation value, and generating a global trade data change trend sequence value; S202: Compare the sequence value of the global trade data change trend with the dynamic trend of the core nodes of the trading network, extract the nodes with significant sequence differences, analyze the dynamic trend of the nodes with significant differences in segments, calculate the mean square error value based on the difference between the node rate and the global dynamic trend rate, and generate the dynamic trend of the core trading nodes; S203: Standard matching is performed on the lag time of the key nodes of the dynamic trend of the core transaction nodes, and the deviation value of the lag time is calculated using the formula: ； Extract nodes with larger lag time deviation values ​​as lag-exceeding nodes, and generate key nodes with lag time exceeding the standard; in, represents the deviation value of the key node lag time, Represents the lag time of the dynamic trend node, represents the standard time series, Represents the total number of nodes, To adjust the weights, the scaling effect of the lag time standard deviation is Represents the standard deviation of the lag time, which measures the dispersion of the time series. Represents the sum operation of all nodes; S204: Analyze the changes in connectivity weights of the key nodes with excessive lag time and the core transaction nodes, calculate connectivity weight fluctuation values ​​based on the node-by-node weight change rate, screen key nodes associated with fluctuation values ​​and lag effects, and establish global key node trend prediction values.

5. The method for digital optimization of trade processes according to claim 4, characterized in that: The steps for obtaining the distributed key node dynamic weight matrix are specifically as follows: S301: Based on the global key node trend prediction value, refine the time change data of each node, extract the time fine-grained changes of multiple nodes through time series decomposition, analyze the change trend in combination with the time point information, and generate time fine-grained node change data; S302: performing time series correlation measurement on the time fine-grained node change data, calculating the time correlation coefficient between adjacent nodes based on the time series of the nodes, selecting nodes with strong correlation through correlation analysis, and generating a time series correlation measurement value for each node; S303: extracting the time change rate in the time series association metric value, calculating the change rate of the node time series by using a differential algorithm, screening and marking nodes with large rate differences, and generating a time change rate value; S304: Compare the time change rate value with the connectivity weight value of the adjacent nodes using the formula: ； Call the node update rules, adjust the weights node by node, and establish a distributed dynamic weight matrix of key nodes; in, represents the updated node weight, represents the original node weight, is the adjustment coefficient, which is used to control the impact of rate changes on weight adjustment. represents the rate difference between nodes, is the average node rate.

6. The method for digital optimization of trade processes according to claim 5, characterized in that: The steps for obtaining the node-associated risk transfer indicator are specifically as follows: S401: extracting the risk factor combination value between nodes based on the distributed key node dynamic weight matrix, analyzing the dynamic change rate of multiple nodes in the weight matrix, and calculating the risk correlation between multiple nodes in combination with the risk factor impact value to generate the risk factor combination value between nodes; S402: Calculate the transmission probability of the risk factor combination value between the nodes, select the node group according to the risk correlation, analyze the transmission path and probability distribution of the risk factor between the nodes, calculate the risk transmission probability in combination with the cumulative transmission model, and generate the risk transmission probability value between the nodes; S403: Compare and analyze the risk transmission probability value between the nodes and the abnormal change area of ​​the weight matrix, and calculate the correlation between the risk transmission probability and the weight change, using the formula: ； The risk transmission probability is weighted based on the weight change rate to screen out risk-significant nodes; in, Represents the association value of the risk-significant node, For the The transmission probability of a node is For the The weight value of each node, is the change in node weight, is the average weight of the nodes, To adjust the impact of changes in weights of factor controls; S404: Analyze the dynamic weight differences of the risk-significant nodes, compare the matching degree between the dynamic weight changes and the transmission probability values ​​between the nodes, calculate the risk transfer relationship between the nodes in combination with the risk correlation, and establish a node-related risk transfer index.

7. The method for digital optimization of trade processes according to claim 6, characterized in that: The steps for obtaining the key node optimization path set are specifically as follows: S501: Based on the node-associated risk transmission index, by analyzing the risk transmission probability and weight change value between nodes one by one, the key nodes with higher transmission probability among the risk nodes are screened, and the initial optimization path is determined in combination with the path flow characteristics to generate an optimized path sequence of the risk nodes; S502: adjusting the key path flow value and weight value of the optimized path sequence of the risk node, calculating the contribution weights of multiple nodes based on the node flow ratio in the path, redistributing the flow value by the total weight difference rate, calculating the adjusted path connectivity index, and adjusting the key path flow value and weight value; S503: Verify the optimized path connectivity after adjusting the critical path flow value and weight value, by calculating the influence of the flow distribution ratio on the path connectivity, using the formula: ； Standardize and optimize the distribution of node traffic on the path, generate optimized traffic distribution based on the weight adjustment ratio, and adjust the traffic distribution ratio; in, Represents the path connectivity index, which measures the impact of traffic distribution and weight between nodes in the path. For the path The traffic value of a node indicates the transmission capacity of the node. is the weight value of the corresponding node, indicating the importance of the node. The total number of nodes on the path, providing a reference for the path size, is an adjustment factor used to control the impact of traffic fluctuations on path connectivity. It represents the standard deviation of the path node traffic and measures the uniformity of node traffic distribution; S504: Optimizing feature analysis is performed on the path after adjusting the traffic distribution ratio, by calculating the traffic balance and node connectivity matching rate within the path, selecting the path according to traffic distribution balance and connectivity, and generating a key node optimized path set according to the optimization rules.

8. A trade process digital optimization system, characterized in that: According to the method for digital optimization of trade processes according to any one of claims 1 to 7, the system comprises: The data feature extraction module extracts the time series feature values ​​of transaction orders, the lag time values ​​of logistics nodes and the amount of abnormal transaction behaviors based on the trade data set, calculates the differences of multiple feature values, and calculates the range of transaction behavior features, compares and adjusts the feature screening rules, optimizes the correlation features between transaction behaviors and logistics nodes, and generates a sparse feature vector set; The global trend analysis module calculates the global trade data change trend sequence value based on the sparse feature vector set, extracts the dynamic change trend of the core transaction nodes, determines whether the key node lag time value meets the standard, analyzes the change of the core transaction node connectivity weight, verifies the consistency, and establishes the global key node trend prediction value; The key node optimization module separates the fine-grained node time change data based on the global key node trend prediction value, calculates the time series correlation measurement value of each node, extracts the time change rate value, compares the connectivity weight values ​​of adjacent nodes, calls the node update rule to iteratively optimize the node correlation, and establishes a distributed key node dynamic weight matrix; The risk factor analysis module extracts the risk factor combination value between nodes based on the distributed key node dynamic weight matrix, calculates the risk transmission probability value, analyzes the abnormal change area in the weight matrix, screens the risk-significant nodes, compares the dynamic weight differences between nodes, and establishes the node-related risk transmission index; The path optimization module constructs an optimized path sequence of risk nodes based on the node-associated risk transfer index, adjusts the key path flow value and weight value, verifies the connectivity of the optimized path, adjusts the flow distribution ratio, and generates a key node optimized path set.

Images