Dynamic configuration method and system based on multi-level virtual account book

Abstract

The application discloses a kind of based on multi-level virtual account book's dynamic configuration method and system, it is related to virtual account book technical field, including obtaining financial transaction preliminary data set and pre-processing, generating feature-scene correlation data;Intelligent hierarchical configuration model is constructed, and preliminary configuration model is generated;Global optimization and local analysis are carried out in clearing path, and intelligent hierarchical analysis model is generated;Based on analysis model evaluation and report generation.The dynamic configuration method and system can improve the efficiency, flexibility and resource utilization of fund management through intelligent hierarchical configuration and global optimization, reduce the redundancy of hierarchical adjustment and fund loss, ensure the accuracy and reliability of fund management scheme, and provide efficient fund management support for different business scenarios.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of virtual account books, in particular to a dynamic configuration method and system based on a multi-level virtual account book. BACKGROUND

[0002] At present, with the continuous upgrading of enterprise financial management needs, the traditional account book system has certain limitations in supporting non-corporate entity classification accounting. At the same time, under the multi-level account book system, the flexibility of hierarchical adjustment is insufficient, and it is difficult to fully meet the real-time business change needs. In addition, the existing account book configuration mechanism lacks intelligent means, especially in transaction clearing optimization and other aspects still needs to be improved. The specific performance is as follows:

[0003] 1. The traditional account book system is mainly designed around the legal entity, and lacks a classification accounting mechanism for non-corporate entities, resulting in the need for member units to open multiple entity accounts through external banks to meet business needs, which not only increases the complexity of account management, but also may lead to scattered storage of funds, reducing the efficiency of fund use.

[0004] 2. The traditional account book system usually adopts a fixed hierarchical structure, and the hierarchical relationship adjustment process is tedious, and it is difficult to quickly respond to changes in business scenarios, especially in the dynamic creation or change of multi-level sub-account books, it is difficult to achieve efficient parameterized configuration, thereby affecting the continuity and adaptability of the business.

[0005] 3. The traditional account book system has limited intelligent degree of configuration corresponding to the transaction clearing path, and when processing transaction data characteristics such as order number, payer information, etc., it still mainly relies on human intervention or simple rule matching, lacks automatic clearing execution capability, and when clearing fails, cannot automatically and timely trigger an early warning or generate a processing work order, further reducing the timeliness and accuracy of account book fund management. SUMMARY

[0006] Therefore, the present application provides a dynamic configuration method and system based on a multi-level virtual account book to solve the technical problems of the limitations of the account book system for non-corporate entities, poor flexibility of hierarchical adjustment of the multi-level account book system, and low intelligent degree of configuration of the transaction clearing path.

[0007] In order to achieve the above purpose, the present application provides the following technical solutions:

[0008] A dynamic configuration method based on a multi-level virtual account book, comprising the following steps:

[0009] Obtaining a financial transaction preparation dataset, performing data preprocessing on the financial transaction preparation dataset to generate a financial transaction preprocessing dataset, continuing to perform transaction feature-business scenario correlation analysis on the financial transaction preprocessing dataset to generate financial transaction feature-scenario correlation data;

[0010] Based on the financial transaction feature-scenario correlation data, an intelligent hierarchical configuration model is constructed to generate a financial transaction hierarchical preliminary configuration model;

[0011] Based on the financial transaction hierarchical preliminary configuration model, a clearing path global optimization is performed to generate a financial transaction global optimization model, and further hierarchical local analysis is performed on the financial transaction global optimization model to generate a financial transaction intelligent hierarchical analysis model;

[0012] Based on the financial transaction intelligent hierarchical analysis model, an evaluation is performed to generate financial transaction evaluation data.

[0013] Based on the above technical solutions, the present application is further described as follows:

[0014] As a further scheme of the present application,

[0015] The data preprocessing on the financial transaction preparation dataset to generate a financial transaction preprocessing dataset specifically includes:

[0016] The financial transaction preparation dataset is subjected to dataset format unification to generate financial transaction unified data, data cleaning is performed according to the financial transaction unified data, and abnormal transaction records are eliminated to generate financial transaction cleaning data, and further data normalization processing is performed on the financial transaction cleaning data to generate a financial transaction preprocessing dataset;

[0017] The financial transaction preprocessing dataset includes a financial transaction feature preprocessing dataset and a financial transaction scenario preprocessing dataset, which are respectively used for transaction feature extraction and business scenario matching analysis.

[0018] As a further scheme of the present application,

[0019] The transaction feature-business scenario correlation analysis on the financial transaction preprocessing dataset to generate financial transaction feature-scenario correlation data specifically includes:

[0020] The financial transaction feature preprocessing dataset is subjected to transaction attribute extraction to generate financial transaction attribute data, and the financial transaction attribute data is subjected to transaction mode analysis to generate financial transaction mode data;

[0021] According to the financial transaction mode data, a transaction network is constructed, and financial transaction network data is generated; business scene matching analysis is performed on the financial transaction network data, and financial transaction scene matching data is generated;

[0022] Transaction frequency detection is performed on the financial transaction scene matching data, and financial transaction frequency data is generated; and transaction amount interval detection is performed on the financial transaction scene matching data, and financial transaction amount interval data is generated;

[0023] Transaction feature calculation is performed on the financial transaction frequency data and the financial transaction amount interval data, and financial transaction feature extraction data is generated;

[0024] According to the financial transaction scene preprocessing data set, transaction risk influence detection is performed, and financial transaction risk data is generated; fund flow direction analysis is performed on the financial transaction risk data, and financial transaction flow direction data is generated; and transaction time distribution measurement is performed on the financial transaction flow direction data, and financial transaction scene feature extraction data is generated;

[0025] The financial transaction feature extraction data and the financial transaction scene feature extraction data are subjected to L2 regularization, and financial transaction feature regularization data is generated;

[0026] According to the financial transaction feature regularization data, a transaction feature-scene feature matrix is created, and financial transaction feature-scene feature matrix is generated; and based on the financial transaction feature-scene feature matrix, transaction feature-scene feature correlation analysis is continued, and financial transaction feature-scene correlation data is generated.

[0027] As a further scheme of the present application,

[0028] The intelligent hierarchical configuration model is constructed based on the financial transaction feature-scene correlation data, and a financial transaction hierarchical preliminary configuration model is generated, specifically including:

[0029] Obtaining financial transaction parameter data provides input constraint conditions for model construction;

[0030] According to the financial transaction feature-scene correlation data, a hierarchical configuration model is differentially constructed, and by analyzing the transaction characteristics in different business scenes, the required account book hierarchical structure of each scene is differentially determined;

[0031] The financial transaction parameter data is used to optimize the algorithm model of the differentially constructed hierarchical configuration model, and a financial transaction hierarchical preliminary configuration model is generated.

[0032] As a further scheme of the present application,

[0033] The financial transaction parameter data is obtained, and input constraint conditions are provided for model construction, specifically including:

[0034] The financial transaction parameter data includes a transaction priority parameter P And a fund flow direction restriction condition C ;

[0035] The algorithm model optimization is performed on the hierarchically configured model constructed in a differentiated manner by using the financial transaction parameter data, and a financial transaction hierarchical preliminary configuration model is generated, specifically including:

[0036] The hierarchical overlap analysis is performed on the hierarchically configured model constructed in a differentiated manner, the financial transaction shared hierarchical data is generated by identifying the hierarchical commonality or shared hierarchy among different scenarios;

[0037] The hierarchical path planning is performed on the financial transaction shared hierarchical data, the optimal fund flow path is calculated by taking the number of hierarchies and processing time consumption as weights, and financial transaction hierarchical path data is generated;

[0038] The path with the least number of hierarchies is taken as the optimal path, when the number of hierarchies is the same, an algorithm is introduced:

[0039] Path= , in the formula, the hierarchical processing time consumption path k is smaller, and the path is better;

[0040] The hierarchical-priority parameter is specified based on the financial transaction hierarchical path data, financial transaction hierarchical parameter data is generated, then according to the efficiency and safety core demand of different business scenarios, the constraint condition corresponding to the financial transaction parameter data is combined, the algorithm model optimization is completed based on the financial transaction hierarchical parameter data and the financial transaction hierarchical path data, and a financial transaction hierarchical preliminary configuration model is generated.

[0041] As a further scheme of the present application,

[0042] The clearing path global optimization is performed based on the financial transaction hierarchical preliminary configuration model, and a financial transaction global optimization model is generated, specifically including:

[0043] The integer linear programming (ILP) algorithm is adopted, and the factors of maximizing the fund utilization rate U and minimizing the clearing path length L are further comprehensively considered, the weight coefficients are integrated into a single optimization objective function, and a financial transaction global optimization model is generated, and a specific formula of the single optimization objective function is:

[0044] Objective function= min ( λ x L+ ( 1−λ ) x ( 1−U ) )

[0045] wherein, λ is a weight coefficient (0≤ λ ≤1), when λ approaches 1, the model pays more attention to the optimization of the clearing path length L , that is, the path with less levels is preferred, when λ approaches 0, the model pays more attention to the improvement of the fund utilization rate U , that is, the idle fund is reduced, and the balance is adjusted flexibly according to the business demand. λ ;

[0046] 1− U converts "maximizing the fund utilization rate" into "minimizing the non-utilization rate", and forms a unified optimization direction with the clearing path length, that is, the global optimization direction is that the objective function is smaller.

[0047] The algorithm formula of the fund utilization rate U is:

[0048]

[0049] In the formula, A is the account node set, N A g max is the maximum available fund of the node g , and the node g represents the starting point, the ending point or the intermediate node of the fund flow, i j a g is the current balance of the node g ; in the above formula, the numerator represents the total idle fund, and the denominator is the total fund, U the closer to 1 indicates that the fund utilization is more sufficient.

[0050] The algorithm formula of the clearing path length L is:

[0051]

[0052] E= {( i, j )} represents the set of all feasible paths, l ij is the level number from the starting point i to the ending point j of the path, x ij is a binary variable, that is, when the path i, j is selected, x ij is 1, otherwise x ij is 0.

[0053] As a further scheme of the present application,​​​

[0054] The hierarchical local analysis of the financial transaction global optimization model generates a financial transaction intelligent hierarchical analysis model, specifically including:

[0055] The clearing path simulation of the financial transaction global optimization model generates financial transaction clearing path simulation data. Then, the financial transaction clearing path simulation data is further optimized by the shortest path generation tree to obtain the financial transaction optimization tree path;

[0056] The optimization tree path is a locally optimal path constructed in the global hierarchical structure for the starting point s and the terminal point t of a specific transaction, in order to achieve local optimization for a specific transaction. The specific formula is as follows:

[0057]

[0058] In the formula, the local analysis node s, t ∈ N , N is the account node set, P st is the optimal path formed from node s to node t , is the set of all feasible paths from node s to node t ;

[0059] l ij is the number of levels between the starting point of a single path i and the terminal point of the path j , i.e. the number of levels of the path i , j ; path ij is the processing time of the path i, j ;

[0060] argmin is used to represent the value of the independent variable that makes the objective function take the minimum value; δ is a conditional switch variable. When the number of levels of different paths is different, δ =0, at this time only the number of levels is calculated, and the processing time is ignored. When the number of levels of different paths is the same, δ =1, since the number of levels is the same, the actual comparison is the processing time;

[0061] The local fund scanning of the financial transaction global optimization model generates financial transaction local fund data, and the specific process is as follows:

[0062] Node gThe formula for calculating idle funds is:

[0063] idle g = A g max - a g

[0064] In the formula, A g max For nodes g The maximum amount of funds available, i.e., the maximum amount of funds that can be used; a g For nodes g The current balance, i.e., the amount of funds that are currently actually owned but not used; idle funds. idle g This refers to the amount of idle funds that can still be utilized after deducting the existing balance, within the maximum usable range.

[0065] The process of generating financial transaction fund optimization data through further optimization of fund utilization is as follows:

[0066] Initial capital utilization rate U The calculation formula is:

[0067]

[0068] By adjusting the allocation of funds, a new balance is generated. a g To minimize the total amount of idle funds after optimization, that is:

[0069]

[0070] The final calculated optimized capital utilization rate is obtained. U '.

[0071] As a further aspect of the present invention,

[0072] The step of performing hierarchical local analysis on the global optimization model of financial transactions to generate an intelligent hierarchical analysis model of financial transactions specifically includes:

[0073] A two-dimensional optimization graph of clearing speed-priority is drawn for the global optimization model of financial transactions, generating a local two-dimensional optimization graph of speed-priority for financial transactions. Based on this optimization graph, a speed-priority matrix is ​​created, and the optimal speed-priority combination is determined, generating the optimal combination data of speed-priority for financial transactions. The specific process is as follows:

[0074] First, establish a clearing speed metric:

[0075] Speed S definition:

[0076] wherein, is the clearing path length;

[0077] Secondly, the comprehensive priority calculation:

[0078] Node g comprehensive priority:

[0079] wherein, P is the transaction priority parameter; α g is the priority coefficient of node g

[0080] Finally, determine the optimal speed-priority matrix combination:

[0081] Define matrix elements: M km = λ ⋅ H k +(1− λ )⋅(1− U m )

[0082] wherein: H k is the level number of the k th path; U m is the fund utilization rate of the m th fund allocation scheme; λ is the weight coefficient (0≤λ≤1);

[0083] Then the optimal combination solving formula is:

[0084]

[0085] The speed optimal solution is , L k is the length of the k th path;

[0086] The priority optimal solution is , α g is the priority coefficient of node g

[0087] As a further scheme of the present application,

[0088] ​​The hierarchical local analysis for the financial transaction global optimization model generates a financial transaction intelligent hierarchical analysis model, and specifically further comprises:

[0089] Based on the financial transaction optimization tree path, the financial transaction re-optimized fund utilization rate data, and the financial transaction speed-priority optimal combination data, the hierarchical local analysis is completed to generate a financial transaction intelligent hierarchical analysis model, and the specific process is:

[0090]

[0091] In the formula, ω 1 + ω 2 + ω 3 = 1, ω 1 is the path optimization weight, affecting the clearing efficiency; ω 2 is the fund utilization rate weight, affecting the resource cost; ω 3 is the speed-priority balance weight, affecting the strategy balance; the specific weight size is determined according to the business scenario, for example, high-frequency transactions are sensitive to clearing speed, and the idle fund cost is low, so ω 1 > ω 2 A dynamic weight configuration calculation formula is specifically introduced as follows:

[0092]

[0093]

[0094] In the formula, A is the current transaction amount, A 0 is the benchmark reference value of the transaction amount; k is the adjustment coefficient, controlling the sensitivity of the weight change, and the value is 0.5; e is the natural constant, and the value is 2.71;

[0095] ω 3 is the balance term, and the value is 0.2.

[0096] A dynamic configuration system according to the dynamic configuration method based on the multi-level virtual account book, the dynamic configuration system comprising:

[0097] The data processing module is configured to obtain a financial transaction preliminary data set, perform data preprocessing on the financial transaction preliminary data set, generate a financial transaction preprocessed data set, continue to perform transaction feature-business scenario correlation analysis on the financial transaction preprocessed data set, and generate financial transaction feature-scenario correlation data.

[0098] The hierarchical configuration model construction module is configured to construct an intelligent hierarchical configuration model based on the financial transaction feature-scenario correlation data, and generate a financial transaction hierarchical preliminary configuration model.

[0099] The optimization and local analysis module is configured to perform global optimization of clearing paths based on the financial transaction hierarchical preliminary configuration model, generate a financial transaction global optimization model, and further perform hierarchical local analysis on the financial transaction global optimization model, and generate a financial transaction intelligent hierarchical analysis model.

[0100] The evaluation and report generation module is configured to perform evaluation based on the financial transaction intelligent hierarchical analysis model, and generate financial transaction evaluation data.

[0101] The present application has the following advantages:

[0102] The present application can realize dynamic management of the financial transaction process, effectively ensures the quality and consistency of data through the data processing module, provides high-quality data support for subsequent modules, effectively completes intelligent hierarchical configuration model construction by means of the hierarchical configuration model construction module, further optimizes the clearing path and the utilization rate of funds by means of the optimization and local analysis module, and provides visual evaluation results through the evaluation and report generation module, thereby significantly improving the overall management flexibility and the efficiency of fund utilization. BRIEF DESCRIPTION OF DRAWINGS

[0103] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed in the following embodiment or prior art description will be briefly introduced. The structure, proportion, size, etc. shown in the present specification are only used to cooperate with the content disclosed in the specification, so as to be understood and read by those skilled in the art. Any modification of structure, change of proportion relationship or adjustment of size, which does not affect the effects and purposes that can be achieved by the present application, should still fall within the scope of the technical content disclosed by the present application.

[0104] Fig. 1 The flowchart of the dynamic management method based on the multi-level virtual account book provided by the embodiment of the present application shows the overall step process from data preprocessing to evaluation report generation.

[0105] Fig. 2The logical structure diagram of the financial transaction level configuration and optimization process in the embodiment of the present application is shown, and the specific implementation process of the level configuration model construction, the global optimization and local analysis of the clearing path are mainly described.

[0106] Fig. 3 The module composition block diagram of the dynamic management system based on the multi-level virtual account book provided by the embodiment of the present application is shown, and the functional principle relationship between the data processing module, the level configuration model construction module, the optimization and local analysis module, and the evaluation and report generation module is shown.

[0107] Fig. 4 The electronic device entity structure schematic diagram of the embodiment of the present application is shown.

[0108] In the drawings, the component list represented by each number is as follows:

[0109] 10, data processing module; 20, level configuration model construction module; 30, optimization and local analysis module; 40, evaluation and report generation module;

[0110] Electronic device 50: processor 501, memory 502, internal bus 503. DETAILED DESCRIPTION

[0111] The embodiments of the present application are described below by specific specific embodiments, and those skilled in the art can easily understand other advantages and effects of the present application from the content disclosed in the specification. Obviously, the described embodiments are part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0112] The terms such as "upper", "lower", "left", "right", "middle" and the like cited in the specification are only for the convenience of clear description, and are not used to limit the scope of the present application. The change or adjustment of the relative relationship is also regarded as the implementable scope of the present application without substantial change of the technical content.

[0113] As Figs. 1 to 3 shown, the embodiment of the present application provides a dynamic configuration method and system based on multi-level virtual account book, and the core is to realize intelligent level configuration, global optimization and local analysis of the financial transaction process through the cooperative work of the data processing module 10, the level configuration model construction module 20, the optimization and local analysis module 30 and the evaluation and report generation module 40. The specific implementation process of the present application is described in detail below in combination with the specific drawings. Figs. 1 to 3

[0114] S1: Fig. 1 ​The overall process of the dynamic management method based on multi-level virtual account books is shown in FIG. 1, which specifically includes the complete steps from data preprocessing to the final generation of evaluation reports. First, the data processing module 10 is responsible for obtaining a financial transaction preliminary data set and performing data preprocessing to ensure the data quality for subsequent analysis.

[0115] Specifically, the data processing module 10 performs operations of steps S11 to S13.

[0116] In step S11, the data processing module 10 obtains a financial transaction preliminary data set by interfacing with an internal financial system or an external transaction interface, which includes but is not limited to transaction amount, transaction timestamp, payer information, payee information, order number, and other original transaction records.

[0117] In step S12, first, the data processing module 10 performs format unification processing on the financial transaction preliminary data set to generate financial transaction unified data, in order to eliminate the difficulty of analysis caused by differences in data formats of different systems, such as converting all date fields to YYYY-MM-DD format and unifying currency units. Then, the data processing module 10 performs data cleaning operations on the financial transaction unified data to eliminate abnormal transaction records such as duplicate records, negative amounts, or missing necessary fields, to generate financial transaction cleaning data. Finally, the data processing module 10 performs normalization processing on the financial transaction cleaning data, which scales the numerical value fields to make the value range of all fields consistent, to generate a financial transaction preprocessing data set, avoiding analysis bias caused by differences in data dimensions.

[0118] The financial transaction preprocessing data set includes a financial transaction feature preprocessing data set and a financial transaction scenario preprocessing data set, which are used for subsequent transaction feature extraction and business scenario matching analysis, respectively.

[0119] In step S13, the data processing module 10 further performs data extraction on the financial transaction preprocessing data set to generate financial transaction extraction data and performs transaction feature and business scenario correlation analysis.

[0120] Specifically, in step S131, first, the data processing module 10 performs transaction attribute data identification extraction on the financial transaction feature pre-processing data set to generate financial transaction attribute data; the transaction attribute data includes transaction types such as transfer payment refund, online and offline transaction channels, and transaction frequency; then, the data processing module 10 performs transaction mode analysis on the financial transaction attribute data to generate financial transaction mode data; more specifically, different transaction modes of high-frequency small-amount transactions and low-frequency large-amount transactions are identified through a clustering algorithm, and based on the different transaction modes, the data processing module 10 constructs financial transaction network data for describing the association between transaction modes and financial transaction network data.

[0121] In step S132, the data processing module 10 performs business scenario matching analysis on the financial transaction network data to generate financial transaction scenario matching data; for example, high-frequency small-amount transactions are matched to retail scenarios, and low-frequency large-amount transactions are matched to bulk procurement scenarios; further, the data processing module 10 performs transaction frequency detection and transaction amount interval detection on the financial transaction scenario matching data to respectively generate financial transaction frequency data and financial transaction amount interval data.

[0122] The financial transaction frequency data and the financial transaction amount interval data are subjected to L2 regularization processing to generate financial transaction feature regularization data for matrix analysis through parameter constraint, feature denoising, and standardization;

[0123] The regularization processing process includes:

[0124] Suppose the transaction frequency data feature is x 1 (unit: times / day), and the transaction amount data feature is x 2 (unit: yuan), and the regularization formula is:

[0125] The normalized feature is

[0126] In the formula, λ is a regularization parameter for controlling the punishment degree; x is substituted into x 1 or x 2.

[0127] Finally, the data processing module 10 creates a transaction feature-scenario feature matrix based on the financial transaction feature regularization data and generates financial transaction feature-scenario association data through matrix analysis, and then uses the financial transaction feature-scenario association data as the data basis for subsequent construction of a hierarchical configuration model.

[0128] S2: Fig. 2 The logic structure of the financial transaction hierarchical configuration and optimization process is shown in FIG.

[0129] The hierarchical configuration model construction module 20 intelligently constructs a hierarchical configuration model based on the financial transaction feature-scene association data, and generates a preliminary financial transaction hierarchical configuration model.

[0130] Specifically, in step S21, the hierarchical configuration model construction module 20 acquires financial transaction parameter data, including transaction priority and fund flow direction restriction conditions, to provide input constraint conditions for model construction.

[0131] More specifically, the transaction priority parameter P and the fund flow direction restriction condition C , P The larger the value, the higher the priority. First, the transaction priority P defines the processing order of different businesses of the model, such as priority of salary payment over ordinary purchase, and the fund flow direction restriction condition C clearly defines business rules, such as defining the model “prohibition of cross-account level transfer” and “daily limit of sub-account”, etc.; second, the transaction priority parameter P is more dependent on high-frequency small-amount transactions in the business scene, in order to ensure processing efficiency, while the fund flow direction restriction condition C is more concerned about low-frequency large-amount transactions, in order to prevent and control fund risk and improve security; in addition, for subsequent hierarchical path planning, the high-priority transaction in the transaction priority parameter P is matched with a shorter fund flow path, and the fund flow direction restriction condition C is used as a constraint condition to clearly exclude illegal paths.

[0132] In step S22, the hierarchical configuration model construction module 20 differentiates the construction of the hierarchical configuration model using the financial transaction feature-scene association data, that is, by analyzing the transaction characteristics in different business scenes, the required account book hierarchical structure of each scene is determined in a differentiated manner; for example, for high-frequency small-amount transaction scenes such as retail, a relatively shallow hierarchical structure is adopted, such as total account → sub-account, a total of 2 layers, to improve processing efficiency; for low-frequency large-amount transaction scenes such as bulk purchase, a relatively deep hierarchical structure is adopted, such as total account → business line account → project account, a total of 3 layers, to strengthen fund control safety, thereby realizing the conversion of business experience into calculable data and realizing the automatic matching of hierarchical structure for different business scenes.

[0133] In step S23, the hierarchical configuration model construction module 20 optimizes the hierarchical configuration model constructed in the above manner, and generates a preliminary financial transaction hierarchical configuration model.

[0134] Specifically, in step S231, the hierarchical configuration model construction module 20 performs hierarchical overlap analysis on the hierarchical configuration model constructed above, i.e., by identifying the hierarchical commonalities or common hierarchies among different scenarios, reducing redundant hierarchical configurations, and finally generating shared hierarchical data of financial transactions.

[0135] More specifically, the intersection of the hierarchical structures of each scenario is identified, for example, "total account" is a common hierarchy for all scenarios, and a common hierarchy set S = {s1, s2, …, s n} is generated, for example, s1= total account.

[0136] In step S232, the hierarchical configuration model construction module 20 performs hierarchical path planning on the shared hierarchical data of financial transactions, specifically, using the Dijkstra shortest path algorithm to calculate the optimal fund flow path with the number of hierarchies and processing time as weights, and generates hierarchical path data of financial transactions.

[0137] More specifically, the path with the least number of hierarchies is taken as the optimal path, and when the number of hierarchies is the same, the formula: path = (1 / hierarchy number) + (1 / processing time) is introduced. , where the hierarchy processing time path k is smaller, the path is better.

[0138] In step S233, the hierarchical configuration model construction module 20 specifies the hierarchy-priority parameter based on the hierarchical path data of financial transactions, generates hierarchical parameter data of financial transactions, for example, where the priority coefficient α : total account α = 1, sub-account α = 0.8, project account α = 0.5;

[0139] Then, according to the efficiency and security core needs of different business scenarios, combined with the constraint conditions corresponding to the financial transaction parameter data, the fund flow restriction conditions C， are avoided, and the final overall priority parameter is determined according to the transaction priority parameter P : P × α ;

[0140] Based on the hierarchical parameter data of financial transactions and the hierarchical path data of financial transactions, the hierarchical configuration model construction module 20 completes the algorithm model optimization and generates a preliminary hierarchical configuration model of financial transactions.

[0141] S3: Based on the preliminary hierarchical configuration model of financial transactions, the clearing path global optimization and hierarchical local analysis are performed by the optimization and local analysis module 30, and a more accurate intelligent hierarchical analysis model is generated.

[0142] In step S31, the optimization and local analysis module 30 performs global optimization of the clearing path based on the preliminary configuration model of the financial transaction hierarchy, transforming the clearing problem into a mathematical model, specifically:

[0143] The Integer Linear Programming (ILP) algorithm is employed to further consider maximizing capital utilization. U Minimize the clearing path length L Factors are integrated into a single optimization objective function through weighting coefficients to generate a global optimization model for financial transactions. The specific formula of the single optimization objective function is as follows:

[0144] Objective function = min ( λ x L+ ( 1−λ ) x ( 1−U ))

[0145] in, λ Weighting coefficients (0≤ λ ≤1), when λ When the value is close to 1, the model pays more attention to the clearing path length. L The optimization, that is, prioritizing the selection of paths with fewer levels, when λ When the value approaches zero, the model focuses more on capital utilization. U The improvement means prioritizing the reduction of idle funds and flexibly adjusting the balance according to business needs. λ ;

[0146] 1− U Transforming "maximizing capital utilization" into "minimizing underutilization" forms a unified optimization direction with the clearing path length; that is, the global optimization direction is to make the objective function smaller.

[0147] Capital utilization rate U The algorithm formula is:

[0148]

[0149] In the formula, N For a set of account nodes, A g max For nodes g Maximum available funds, nodes g Represents the starting point of capital flow i ,end j Or intermediate nodes, a g For nodes g The current balance; in the above formula, the numerator represents the total amount of idle funds, and the denominator is the total funds. U The closer to 1, the more fully the funds are utilized.

[0150] Clearing path length L The algorithm formula is:

[0151]

[0152] E= {( i, j )} represents the set of all feasible paths. l ij The starting point of a single path i To the end of the route j The number of levels between them x ij It is a binary variable, namely the path selection ( i, j )hour, x ij =1, otherwise x ij It is 0.

[0153] In step S32, the optimization and local analysis module 30 further performs hierarchical local analysis on the global optimization model of financial transactions to generate an intelligent hierarchical analysis model of financial transactions.

[0154] Specifically, in step S321, the optimization and local analysis module 30 simulates the clearing path of the global optimization model for financial transactions, generating simulated clearing path data for financial transactions. Then, it further optimizes the simulated clearing path data for financial transactions using a shortest path spanning tree to obtain an optimized tree path for the financial transactions. This optimized tree path is a locally optimal path constructed within the global hierarchical structure for a specific transaction's starting point s and ending point t, aiming to achieve local optima for that specific transaction. The specific formula is as follows:

[0155]

[0156] In the formula, local analysis nodes s, t ∈ N , N For a set of account nodes, P st For the node s To the node t The optimal path formed For nodes s To the node t The set of all feasible paths;

[0157] l ij The starting point of a single path i To the end of the route j The number of levels between them, i.e., the path ( i , j The number of levels; pathij The processing time of the path (T i, j );

[0158] argmin The value of the independent variable that makes the objective function minimum; δ The conditional switch variable, when the number of levels of different paths is different, δ =0, at this time only the number of levels is calculated, ignoring the processing time, and when the number of levels of different paths is the same, δ =1, since the number of levels is the same, the actual comparison is the processing time.

[0159] In step S322, the optimization and local analysis module 30 performs local fund scanning on the global financial transaction optimization model to generate financial transaction local fund data, and the specific process is as follows:

[0160] The idle fund calculation formula of the node g is as follows:

[0161] idle g = A g max − a g

[0162] In the formula, A g max The maximum available fund of the node g , that is, the maximum amount of fund that can be used; a g The current balance of the node g , that is, the actual amount of fund that is currently owned and not used; the idle fund idle g is the amount of idle fund that can be used within the maximum available range after the existing balance is removed.

[0163] The financial transaction fund re-optimization data is generated by re-optimizing the fund utilization rate, and the specific process is as follows:

[0164] The calculation formula of the initial fund utilization rate U is as follows:

[0165]

[0166] By adjusting the fund allocation, a new balance a g is generated, so that the total idle fund after optimization is minimized, that is:

[0167]

[0168] The final calculation obtains the re-optimized capital utilization rate U

[0169] In step S323, the optimization and local analysis module 30 performs a clearing speed-priority two-dimensional optimization mapping for the global optimization model of the financial transaction, generates a local speed-priority two-dimensional optimization map of the financial transaction, and based on the optimization map, the optimization and local analysis module 30 creates a speed-priority matrix and determines the optimal speed-priority combination, generates financial transaction speed-priority optimal combination data, the specific process is:

[0170] First, construct the clearing speed index:

[0171] Speed S definition:

[0172] In the formula, is the clearing path length;

[0173] Second, comprehensive priority calculation:

[0174] The comprehensive priority of node g :

[0175] In the formula, P is the transaction priority parameter; α g is the priority coefficient of node g , such as the total account α=1.

[0176] Finally, determine the optimal speed-priority matrix combination:

[0177] Define the matrix element: M km = λ ⋅ H k +(1− λ )⋅(1− U m )

[0178] In the formula: H k is the level number of the first k path; U m is the capital utilization rate of the first m capital allocation scheme; λ is the weight coefficient (0≤λ≤1).

[0179] Then the optimal combination solving formula is:

[0180]

[0181] ​The speed optimal solution is , L k The length of the first path; k

[0182] The priority optimal solution is , α g The priority coefficient of the node g ;

[0183] Finally, in step S324, the optimization and local analysis module 30 completes the hierarchical local analysis based on the financial transaction optimization tree path, the financial transaction re-optimized fund utilization rate data, and the financial transaction speed-priority optimal combination data, generates a financial transaction intelligent hierarchical analysis model, and the specific process is as follows:

[0184]

[0185] In the formula, ω 1 + ω 2 + ω 3 =1, ω 1 The path optimization weight affects the clearing efficiency; ω 2 The fund utilization rate weight affects the resource cost; ω 3 The speed-priority balance weight affects the strategy balance. The specific weight size is determined according to the business scenario. For example, high-frequency transactions are sensitive to clearing speed, and the idle fund cost is low, so ω 1 ω 2 A dynamic weight configuration calculation formula is introduced as follows:

[0186]

[0187]

[0188] In the formula, A is the current transaction amount, A 0 is the benchmark reference value of the transaction amount, and the value in the embodiment is 100,000; k is an adjustment coefficient that controls the sensitivity of weight change, and the value is 0.5; e is a natural constant, and the value is 2.71; ω 3 is a balance term, and the value is 0.2.

[0189] When A k(A−A0) ​​≈ 0, at this time ω1≈ 1;

[0190] When A = A0, e k(A−A0) = e 0 = 1, at this time ω1= 0.5;

[0191] When A > A0, e k(A−A0) → +∞, at this time ω1≈ 0.

[0192] S4: The evaluation and report generation module 40 evaluates and generates an evaluation report based on the financial transaction intelligent hierarchical analysis model.

[0193] In step S41, the evaluation and report generation module 40 acquires real-time financial transaction data, including the current transaction amount, transaction time and payer information.

[0194] In step S42, the evaluation and report generation module 40 performs financial transaction operations based on the financial transaction intelligent hierarchical analysis model to generate financial transaction model data; then the evaluation and report generation module 40 compares the real-time financial transaction data with the financial transaction model data to generate financial transaction evaluation data.

[0195] In step S43, the evaluation and report generation module 40 generates an evaluation report for the financial transaction evaluation data, thereby completing the dynamic management operation based on the multi-level virtual account book; this process can provide intuitive fund management suggestions for enterprise managers by visualizing the evaluation results.

[0196] Through the cooperative work of the above modules, the application can realize dynamic management of the financial transaction process, effectively ensure the quality and consistency of the data through the data processing module 10, provide high-quality data support for subsequent modules, and improve the intelligent level of hierarchical configuration by using the hierarchical configuration model construction module 20, in addition, the optimization and local analysis module 30 can be used to optimize the clearing path and fund utilization rate, and the evaluation and report generation module 40 can be used to provide visual evaluation results, thereby improving the overall management and control efficiency.

[0197] As shown in Fig. 4 The embodiment of the application also provides a dynamic configuration system according to the dynamic configuration method based on the multi-level virtual account book, and the dynamic configuration system comprises:

[0198] The data processing module 10 is used for acquiring a financial transaction preliminary data set, performing data preprocessing on the financial transaction preliminary data set to generate a financial transaction preprocessed data set, and continuing to perform transaction feature-business scene correlation analysis on the financial transaction preprocessed data set to generate financial transaction feature-scene correlation data.

[0199] The hierarchical configuration model construction module 20 is configured to construct an intelligent hierarchical configuration model based on the financial transaction feature-scenario association data, and generate a financial transaction hierarchical preliminary configuration model.

[0200] The optimization and local analysis module 30 is configured to perform global optimization of the clearing path based on the financial transaction hierarchical preliminary configuration model, generate a financial transaction global optimization model, and further perform hierarchical local analysis on the financial transaction global optimization model, and generate an intelligent hierarchical analysis model of the financial transaction.

[0201] The evaluation and report generation module 40 is configured to perform evaluation based on the intelligent hierarchical analysis model of the financial transaction, and generate financial transaction evaluation data.

[0202] Fig. 4 An electronic device entity structure schematic diagram of an embodiment of the present application is shown in FIG. 5. Fig. 4 As shown in FIG. 5, the electronic device 50 includes a processor 501, a memory 502, and an internal bus 503; wherein the processor 501 and the memory 502 communicate with each other through the internal bus 503.

[0203] The processor 501 is configured to invoke program instructions in the memory 502 to execute the method provided by each method embodiment described above, for example, including: obtaining a financial transaction preliminary data set, performing data preprocessing on the financial transaction preliminary data set to generate a financial transaction preprocessed data set, continuing to perform transaction feature-business scenario association analysis on the financial transaction preprocessed data set to generate financial transaction feature-scenario association data; constructing an intelligent hierarchical configuration model based on the financial transaction feature-scenario association data, and generating a financial transaction hierarchical preliminary configuration model; performing global optimization of the clearing path based on the financial transaction hierarchical preliminary configuration model, generating a financial transaction global optimization model, and further performing hierarchical local analysis on the financial transaction global optimization model, and generating an intelligent hierarchical analysis model of the financial transaction; and performing evaluation based on the intelligent hierarchical analysis model of the financial transaction, and generating financial transaction evaluation data.

[0204] The embodiment provides a non-transitory computer readable storage medium storing computer instructions, the computer instructions causing a computer to execute the method provided by each method embodiment, for example comprising: obtaining a financial transaction preparation data set, performing data preprocessing on the financial transaction preparation data set, generating a financial transaction preprocessing data set, continuing to perform transaction feature-business scenario correlation analysis on the financial transaction preprocessing data set, and generating financial transaction feature-scenario correlation data; performing intelligent hierarchical configuration model construction based on the financial transaction feature-scenario correlation data, generating a financial transaction hierarchical preliminary configuration model; performing clearing path global optimization based on the financial transaction hierarchical preliminary configuration model, generating a financial transaction global optimization model, and further performing hierarchical local analysis on the financial transaction global optimization model, generating a financial transaction intelligent hierarchical analysis model; performing evaluation based on the financial transaction intelligent hierarchical analysis model, and generating financial transaction evaluation data.

[0205] Those skilled in the art can understand that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware, and the foregoing program can be stored in a computer readable storage medium. When the program is executed, the steps of the method embodiments are executed. The foregoing storage medium includes ROM, RAM, magnetic disc or optical disc and various storage media that can store program codes.

[0206] The device embodiments described above are only schematic, and units described as separate components can or can not be physically separate, and components shown as units can or can not be physical units, that is, they can be located in one place, or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0207] From the above description of the embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and necessary universal hardware platforms, and of course, can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of software products, which can be stored in a computer readable storage medium, such as ROM / RAM, magnetic disc, optical disc, etc., and include a plurality of instructions to make a computer device (which can be a server, a network device, etc.) execute the method of each embodiment or some part of the embodiment.

[0208] Although the present application has been described in detail with general description and specific embodiments above, it is obvious to those skilled in the art that some modifications or improvements can be made on the basis of the present application. Therefore, these modifications or improvements made on the basis of not deviating from the spirit of the present application, all belong to the scope of protection claimed by the present application.

Claims

1. A dynamic configuration method based on a multi-level virtual ledger, characterized in that, Includes the following steps: Obtain a preliminary financial transaction dataset, perform data preprocessing on the preliminary financial transaction dataset to generate a preprocessed financial transaction dataset, and then perform transaction feature-business scenario correlation analysis on the preprocessed financial transaction dataset to generate financial transaction feature-scenario correlation data. An intelligent hierarchical configuration model is constructed based on financial transaction characteristics and scenario-related data, generating a preliminary configuration model for financial transaction hierarchies. Based on the preliminary configuration model of the financial transaction hierarchy, the clearing path is globally optimized to generate a global optimization model of financial transactions. Furthermore, a hierarchical local analysis is performed on the global optimization model of financial transactions to generate an intelligent hierarchical analysis model of financial transactions. The evaluation is conducted based on an intelligent hierarchical analysis model for financial transactions, generating financial transaction evaluation data. The step of performing global optimization of the clearing path based on the preliminary configuration model of the financial transaction hierarchy to generate a global optimization model for financial transactions specifically includes: An integer linear programming algorithm is employed to further consider maximizing capital utilization. U Minimize the clearing path length L Factors are integrated into a single optimization objective function through weighting coefficients to generate a global optimization model for financial transactions. The specific formula of the single optimization objective function is as follows: Objective function = min ( λ x L+ ( 1−λ ) x ( 1−U )) in, λ For weighting coefficients, 0 ≤ λ ≤1, when λ When the value is close to 1, the model pays more attention to the clearing path length. L The optimization, that is, prioritizing the selection of paths with fewer levels, when λ When the value is close to 0, the model focuses more on capital utilization. U The improvement means prioritizing the reduction of idle funds and flexibly adjusting the balance according to business needs. λ ; 1− U Transforming "maximizing capital utilization" into "minimizing underutilization" forms a unified optimization direction with the clearing path length; that is, the global optimization direction is to make the objective function smaller. Capital utilization rate U The algorithm formula is: In the formula, N For a set of account nodes, A g max For nodes g Maximum available funds, nodes g Represents the starting point of capital flow i ,end j Or intermediate nodes, a g For nodes g The current balance; in the above formula, the numerator represents the total amount of idle funds, and the denominator is the total funds. U The closer it is to 1, the more fully the funds are utilized; Clearing path length L The algorithm formula is: E= {( i, j )} represents the set of all feasible paths. l ij The starting point of a single path i To the end of the route j The number of levels between them x ij It is a binary variable, namely the path selection ( i, j )hour, x ij =1, otherwise x ij =0; The step of performing hierarchical local analysis on the global optimization model of financial transactions to generate an intelligent hierarchical analysis model of financial transactions specifically includes: The clearing path is simulated for the global optimization model of financial transactions to generate financial transaction clearing path simulation data; then, the shortest path spanning tree is further optimized for the financial transaction clearing path simulation data to obtain the financial transaction optimized tree path. Optimized tree path is a locally optimal path constructed within the global hierarchical structure for a specific transaction's starting point s and ending point t, aiming to achieve local optima for that specific transaction. The specific formula is as follows: In the formula, local analysis nodes s,t ∈ N , N For a set of account nodes, P st For the node s To the node t The optimal path formed For nodes s To the node t The set of all feasible paths; l ij The starting point of a single path i To the end of the route j The number of levels between them, i.e., the path ( i , j The number of levels; path ij For path ( i, j The processing time of ) argmin Used to represent the value of the independent variable that minimizes the objective function; δ This is a conditional switch variable; it is used when the number of levels differs across different paths. δ =0, in this case only the number of levels is calculated and the processing time is ignored; however, when the number of levels is the same for different paths, then... δ =1, since the number of levels is the same, what is actually being compared is the processing time; Continue by performing a local fund scan on the aforementioned global financial transaction optimization model to generate local financial transaction fund data. The specific process is as follows: node g The formula for calculating idle funds is: idle g = A g max − a g In the formula, A g max For nodes g The maximum amount of funds available, i.e., the maximum amount of funds that can be used; a g For nodes g The current balance, i.e., the amount of funds that are currently actually owned but not used; idle funds. idle g This refers to the amount of idle funds that can still be utilized after deducting the existing balance, within the maximum usable range. The process of generating financial transaction fund optimization data through further optimization of fund utilization is as follows: Initial capital utilization rate U The calculation formula is: By adjusting the allocation of funds, a new balance is generated. a g To minimize the total amount of idle funds after optimization, that is: The final calculated optimized capital utilization rate is obtained. U '.

2. The dynamic configuration method based on a multi-level virtual ledger according to claim 1, characterized in that, The step of preprocessing the financial transaction preprocessing dataset to generate a financial transaction preprocessing dataset specifically includes: The dataset format of the financial transaction preparation dataset is standardized to generate standardized financial transaction data. The standardized financial transaction data is then cleaned to remove abnormal transaction records, generating cleaned financial transaction data. The cleaned financial transaction data is then normalized to generate a preprocessed financial transaction dataset. The financial transaction preprocessing dataset includes a financial transaction feature preprocessing dataset and a financial transaction scenario preprocessing dataset, which are applied one-to-one to transaction feature extraction and business scenario matching analysis, respectively.

3. The dynamic configuration method based on a multi-level virtual ledger according to claim 2, characterized in that, The step of performing transaction feature-business scenario correlation analysis on the preprocessed financial transaction dataset to generate financial transaction feature-scenario correlation data specifically includes: Transaction attributes are extracted from the preprocessed dataset of financial transaction features to generate financial transaction attribute data; transaction pattern analysis is then performed on the financial transaction attribute data to generate financial transaction pattern data. A transaction network is constructed based on financial transaction pattern data to generate financial transaction network data; business scenario matching analysis is performed on the financial transaction network data to generate financial transaction scenario matching data. Transaction frequency detection is performed on the financial transaction scenario matching data to generate financial transaction frequency data, and transaction amount range detection is performed on the financial transaction scenario matching data to generate financial transaction amount range data. Financial transaction frequency data and financial transaction amount range data are used to calculate transaction characteristics and generate financial transaction characteristic extraction data. Based on the preprocessed dataset of the financial transaction scenario, transaction risk impact detection is performed to generate financial transaction risk data. Fund flow analysis is performed on the financial transaction risk data to generate financial transaction flow data. Transaction time distribution is measured on the financial transaction flow data to generate financial transaction scenario feature extraction data. The financial transaction feature extraction data and the financial transaction scenario feature extraction data are subjected to L2 regularization to generate financial transaction feature regularization data. A transaction feature-scenario feature matrix is ​​created based on the regularized data of financial transaction features. Then, a transaction feature-scenario feature correlation analysis is performed based on the financial transaction feature-scenario feature matrix to generate financial transaction feature-scenario correlation data.

4. The dynamic configuration method based on a multi-level virtual ledger according to claim 3, characterized in that, The construction of an intelligent hierarchical configuration model based on financial transaction characteristics and scenario-related data, generating a preliminary configuration model for financial transaction hierarchies, specifically includes: Obtain financial transaction parameter data to provide input constraints for model construction; Based on the financial transaction characteristics and scenario-related data, a hierarchical configuration model is constructed in a differentiated manner. By analyzing the transaction characteristics under different business scenarios, the required ledger hierarchy structure for each scenario is determined accordingly. The algorithm model is optimized using the financial transaction parameter data to construct a hierarchical configuration model with differentiation, thereby generating a preliminary configuration model for the financial transaction hierarchy.

5. The dynamic configuration method based on a multi-level virtual ledger according to claim 4, characterized in that, The acquisition of financial transaction parameter data, which provides input constraints for model construction, specifically includes: The financial transaction parameter data includes transaction priority parameters. P and restrictions on the flow of funds C ; The step of optimizing the algorithm model of the differentiated hierarchical configuration model using the financial transaction parameter data to generate a preliminary configuration model of the financial transaction hierarchy specifically includes: For the hierarchical configuration model constructed differently, hierarchical overlap analysis is performed. By identifying the common or shared hierarchical levels between different scenarios, financial transaction shared hierarchical data is generated. Hierarchical path planning is performed on the shared hierarchical data of financial transactions. The optimal fund flow path is calculated with the number of levels and processing time as weights, and financial transaction hierarchical path data is generated. The path with the fewest levels is selected as the optimal path. When the number of levels is the same, a formula is introduced: Path = The time consumed by the hierarchical processing in the formula path k Smaller is a better path; Based on the financial transaction hierarchy path data, hierarchy-priority parameters are specified to generate financial transaction hierarchy parameter data. Then, according to the efficiency and security core requirements of different business scenarios, combined with the constraints corresponding to the financial transaction parameter data, the algorithm model is optimized based on the financial transaction hierarchy parameter data and the financial transaction hierarchy path data to generate a preliminary configuration model for the financial transaction hierarchy.

6. The dynamic configuration method based on a multi-level virtual ledger according to claim 5, characterized in that, The step of performing hierarchical local analysis on the global optimization model of financial transactions to generate an intelligent hierarchical analysis model of financial transactions specifically includes: A two-dimensional optimization graph of clearing speed-priority is drawn for the global optimization model of financial transactions, generating a local two-dimensional optimization graph of speed-priority for financial transactions. Based on this optimization graph, a speed-priority matrix is ​​created, and the optimal speed-priority combination is determined, generating the optimal combination data of speed-priority for financial transactions. The specific process is as follows: First, establish a clearing speed metric: Velocity S is defined as follows: In the formula, To clarify path length; Secondly, the overall priority calculation: node g Overall priority: In the formula, P For transaction priority parameters; α g For nodes g Priority coefficient; Finally, determine the optimal speed-priority matrix combination: Define matrix elements: M km = λ ⋅ H k +(1− λ )⋅(1− U m ) In the formula: H k For the first k The number of levels in the path; U m For the first m The capital utilization rate of various capital allocation schemes; λ The weighting coefficient is 0 ≤ λ ≤ 1; The formula for finding the optimal combination is: The optimal solution for speed is , L k For the first k The length of the path; The optimal solution with priority is , α g For nodes g Priority coefficient.

7. The dynamic configuration method based on a multi-level virtual ledger according to claim 6, characterized in that, The step of performing hierarchical local analysis on the global optimization model of financial transactions to generate an intelligent hierarchical analysis model of financial transactions specifically includes: Based on the optimized tree path of financial transactions, the capital utilization rate data after further optimization of financial transactions, and the optimal combination data of financial transaction speed-priority, hierarchical local analysis is completed to generate an intelligent hierarchical analysis model for financial transactions. ω 1 , ω 2 , ω 3 ,in ω 1 Optimizing path weights affects sorting efficiency; ω 2 The weighting of capital utilization rate affects resource costs; ω 3 The speed-priority balancing weights affect the strategy's balance; the specific weights are determined based on the business scenario. For high-frequency trading pairs that are sensitive to clearing speed and have low idle capital costs, a certain weight is used. ω 1 > ω 2 The specific formula for calculating a dynamic weight configuration is as follows: In the formula, A This is the current transaction amount. A 0 This serves as a benchmark reference value for the transaction amount; k is the adjustment coefficient, which controls the sensitivity of weight changes, and is set to 0.5; e is the natural constant, and is set to 2.

71. ω 3 It is a balancing term, with a value of 0.

2.

8. A dynamic configuration system based on the dynamic configuration method of a multi-level virtual ledger as described in claim 7, characterized in that, The dynamic configuration system includes: The data processing module is used to acquire the financial transaction preparatory dataset, perform data preprocessing on the financial transaction preparatory dataset to generate the financial transaction preprocessed dataset, and further perform transaction feature-business scenario correlation analysis on the financial transaction preprocessed dataset to generate financial transaction feature-scenario correlation data. The hierarchical configuration model building module is used to build an intelligent hierarchical configuration model based on financial transaction characteristics and scenario-related data, and generate a preliminary hierarchical configuration model for financial transactions. The optimization and local analysis module is used to perform global optimization of the clearing path based on the initial configuration model of the financial transaction hierarchy, generate a global optimization model of financial transactions, and further perform hierarchical local analysis on the global optimization model of financial transactions to generate an intelligent hierarchical analysis model of financial transactions. The assessment and report generation module is used to conduct assessments based on an intelligent hierarchical analysis model for financial transactions and generate financial transaction assessment data.

Images