Data simulation method and device, computer equipment, readable storage medium and program product

By receiving and analyzing historical flow data, predicting future flow data, and simulating market interactions, the problem of inaccurate simulation results in the existing technology is solved, and more accurate market forecasting and decision-making support is achieved.

CN118941320AInactive Publication Date: 2024-11-12JIANZHI DATA TECH (SHANGHAI) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411125390.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-16
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing enterprise operation simulation software can only simulate the company's decision-making scenarios separately, and does not consider the company's interaction in the market environment, resulting in inaccurate simulation results.

Method used

By receiving historical flow data, predicting future flow data, and querying interactive objects in historical market flow data, combining future flow data and market data, simulate future flow data between objects in the target market, considering multiple time periods and complexity Interactive relationship.

Benefits of technology

It enhances prediction accuracy, dynamically reflects market changes, and the generated future flow data is more accurate, taking into account the complex interactive relationships in the market.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118941320A_ABST
    Figure CN118941320A_ABST
Patent Text Reader

Abstract

The invention relates to a data simulation method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: receiving historical pipeline data between a first object and a second object; according to the historical flow data, predicting first future flow data between the first object and the second object; querying a reference object interacting with the first object and the second object in historical market flow data; simulating second future flow data among objects in a target simulation market according to the first future flow data and the historical market flow data to obtain a data simulation result; the target simulation market comprises the first object, the second object and the reference object. Accurate simulation can be carried out by adopting the method.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a data simulation method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] With the rapid development of artificial intelligence technology, especially large model technology, combining AI with corporate decision-making to help companies formulate more optimized and forward-looking business strategies has become a new technological trend. At the same time, for companies, how to try and optimize decision-making plans in a risk-free environment, thereby reducing business risks in the actual implementation process, has also received increasing attention.

[0003] Therefore, some business simulation software based on historical data has emerged to simulate the operating results of enterprises under different decision-making scenarios to assist enterprises in making decisions. However, the current simulation method only simulates the enterprise as an individual, and does not consider the interaction between the enterprise and other enterprises in the market environment, which leads to inaccurate simulation results. Summary of the invention

[0004] Based on this, it is necessary to provide a data simulation method, device, computer equipment, computer-readable storage medium and computer program product that can accurately simulate the above-mentioned technical problems.

[0005] In a first aspect, the present application provides a data simulation method, the method comprising:

[0006] Receiving historical flow data between the first object and the second object;

[0007] Predicting first future pipeline data between the first object and the second object based on the historical pipeline data;

[0008] Querying reference objects that interact with the first object and the second object in historical market flow data;

[0009] According to the first future flow data and the historical market flow data, the second future flow data between the objects in the target simulation market is simulated to obtain a data simulation result; the target simulation market includes the first object, the second object and the reference object.

[0010] In one embodiment, the first future pipeline data and the second future pipeline data include first sub-pipeline data and second sub-pipeline data corresponding to a plurality of sub-time periods;

[0011] The step of simulating the second future flow data between the objects in the target simulation market according to the first sub-flow data and the historical market flow data to obtain a data simulation result includes:

[0012] According to the first sub-flow data and historical market flow data corresponding to different sub-time periods, the second sub-flow data between various objects in the target simulation market corresponding to different sub-time periods are simulated to obtain the data simulation result.

[0013] In one embodiment, simulating the second sub-flow data between various objects in the target simulation market corresponding to different sub-time periods according to the first sub-flow data corresponding to different sub-time periods and the historical market flow data to obtain the data simulation result includes:

[0014] According to the first sub-flow data corresponding to the current sub-time period and the historical market flow data corresponding to the previous sub-time period, simulating the second sub-flow data between various objects in the target simulation market corresponding to the current sub-time period;

[0015] The historical market flow data is updated based on the second sub-flow data, and according to the updated historical market flow data and the first sub-flow data corresponding to the next sub-time period, the step of simulating the second sub-flow data corresponding to the next sub-time period is continued until the data simulation result is obtained.

[0016] In one embodiment, the updating of the historical market transaction data based on the second sub-transaction data includes:

[0017] Integrate the second sub-flow data with the historical market flow data to obtain intermediate reference flow data;

[0018] The target market flow data is extracted from the intermediate reference flow data, and the target market flow data is used as the updated historical market flow data.

[0019] In one embodiment, extracting target market flow data from intermediate reference flow data includes:

[0020] Calculate the recency index, importance index and relevance index of each transaction data in the historical market transaction data;

[0021] Calculate a comprehensive score for each of the flow data by comprehensively considering the recency index, the importance index, and the relevance index;

[0022] The target market flow data is extracted from the intermediate reference flow data according to the comprehensive score to obtain the target market flow data.

[0023] In one embodiment, simulating second future flow data between various objects in the target simulation market according to the first sub-flow data and the historical market flow data to obtain a data simulation result further includes:

[0024] Acquiring simulation parameters; the simulation parameters include one or more of enterprise operation strategy, competitor strategy, external environment parameters, and object attributes;

[0025] According to the simulation parameters, the first sub-flow data and the historical market flow data, the first object and the second future flow data between the second object and the reference object are simulated to obtain the data simulation result.

[0026] In one embodiment, the prediction of the first sub-flow data between the first object and the second object based on the historical flow data is obtained by a pre-trained flow prediction model; the prediction process of the flow prediction model includes:

[0027] Acquire sample flow data; the sample flow data carries a flow tag;

[0028] Inputting the sample flow data into the initial model for training to obtain the predicted flow;

[0029] A target loss function is obtained based on the predicted flow and the flow label, and the parameters of the initial model are adjusted according to the target loss function until the training is completed to obtain the flow prediction model.

[0030] In one embodiment, before the sample pipeline data is input into the initial model for training, the method further includes:

[0031] Preprocessing the sample flow data;

[0032] The preprocessing includes one or more of anonymization and cleaning.

[0033] In one embodiment, the method further comprises:

[0034] Get the objective function and constraints;

[0035] Based on the objective function and the constraints, a target decision plan corresponding to the data simulation result is calculated.

[0036] In a second aspect, the present application provides a data simulation device, the device comprising:

[0037] A receiving module, used for receiving historical flow data between the first object and the second object;

[0038] A prediction module, configured to predict first future pipeline data between the first object and the second object based on the historical pipeline data;

[0039] A query module, used to query reference objects that interact with the first object and the second object in historical market flow data, and obtain the reference objects;

[0040] A simulation module is used to simulate the second future flow data between various objects in the target simulation market according to the first future flow data and the historical market flow data to obtain data simulation results; the target simulation market includes the first object, the second object and the reference object.

[0041] In a third aspect, the present application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method in any one of the above embodiments when executing the computer program.

[0042] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method in any one of the above embodiments are implemented.

[0043] In a fifth aspect, the present application provides a computer program product, including a computer program, which implements the steps of the method in any one of the above embodiments when executed by a processor.

[0044] The above-mentioned data simulation method, device, computer equipment, computer-readable storage medium and computer program product, by receiving the historical flow data between the first object and the second object, and then predicting the first future flow data based on the historical flow data, can fully understand the past transaction patterns and behaviors. Secondly, by querying the reference objects that interact with the first object and the second object in the historical market flow data, other objects that have a significant impact on market transactions can be identified, thereby simulating the market environment more comprehensively. Finally, combining the first future flow data and the historical market flow data, the second future flow data between the various objects in the target simulation market is simulated, and the transaction behavior between the various objects in the future time period is simulated, which can reflect the dynamic interaction between the various objects. In this way, not only the prediction accuracy is enhanced, but also the complex interactive relationship in the market is taken into account, and the market changes can be dynamically reflected, so that the generated second future flow data is more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0046] Figure 1 A schematic diagram of a flow chart of a data simulation method in one embodiment;

[0047] Figure 2 A schematic diagram of a flow chart of a data simulation method in another embodiment;

[0048] Figure 3 is a structural block diagram of a data simulation device in one embodiment;

[0049] Figure 4 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application 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 application and are not used to limit the present application.

[0051] In one embodiment, Figure 1 As shown, a data simulation method is provided. This embodiment uses the method applied to a terminal as an example for illustration. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0052] Step S102, receiving historical flow data between the first object and the second object:

[0053] The object can be any entity that needs to interact, which can be an enterprise, an individual consumer, a financial institution, a node in a supply chain, etc.

[0054] For example, if the object is an enterprise, the historical transaction data is the various cash flow records of the enterprise over a period of time, which can be sales revenue records, purchase records, operating expenditure records, and production costs, etc. These records reflect the financial activities and operating conditions of the enterprise.

[0055] For example, if the object is a node in a supply chain, the historical flow data may be raw material procurement records, product inventory records, logistics transportation records, and the like.

[0056] Step S104: predicting first future flow data between the first object and the second object based on the historical flow data.

[0057] The first future flow data refers to the flow data of the first object and the second object in a period of time in the future.

[0058] Optionally, the first future flow data may be predicted by a pre-trained flow prediction model, an autoregressive moving average model, a rolling average method, a Monte Carlo simulation, or the like.

[0059] Exemplarily, the historical flow data may be input into a pre-trained flow prediction model, and the first future flow data may be predicted by the flow prediction model.

[0060] Exemplarily, when using Monte Carlo prediction, it is assumed that the historical flow data follows a normal distribution or other suitable distribution, but in this embodiment, the normal distribution is taken as an example. According to the assumed distribution, according to the statistical characteristics of the historical data, such as the mean and standard deviation, random sampling is performed to generate multiple possible future values, and multiple repeated experiments are performed. Through a large number of predictions, a set of possible future values ​​can be obtained. Finally, all possible future values ​​are statistically analyzed, such as calculating the mean, standard deviation, and confidence interval, etc., to obtain the first future flow data.

[0061] Step S106, querying reference objects that interact with the first object and the second object in the historical market flow data.

[0062] Among them, historical market flow data refers to the detailed record of transaction activities between all parties in market transactions. Historical market flow data usually includes the time of transaction, participants, transaction amount, transaction method and other relevant information.

[0063] Optionally, a first object and a second object are determined, for example, the first object is object A, and the second object is object B, and then all transaction data related to object A and object B are extracted from historical market transaction data, and these transaction data include transactions sent or received by object A or transactions sent or received by object B, and these transaction data are used as reference transaction data. Then, all objects that have transaction interactions with object A and all objects that have transaction interactions related to object B are found from the reference transaction data, and objects that have transaction interactions with both object A and object B are used as reference objects.

[0064] Furthermore, the reference object includes, but is not limited to, an object that directly transacts with object A or object B, and may also include an object that indirectly transacts with object A or object B.

[0065] Step S108, simulating the second future flow data between the objects in the target simulation market according to the first future flow data and the historical market flow data to obtain a data simulation result; the target simulation market includes the first object, the second object and the reference object.

[0066] The target simulation market is composed of a first object, a second object and a reference object. In the target simulation market, each object interacts with each other. In this embodiment, a transaction between objects is regarded as an interaction between objects.

[0067] The second future flow data refers to the flow data generated by the interaction between various objects in the simulated market in a future period of time. It should be noted that the time period corresponding to the first future flow needs to be consistent with the time period corresponding to the second future flow.

[0068] Optionally, in this embodiment, the predicted first future flow between the first object and the second object and the known historical market flow are used to simulate a period of time in the future to simulate second future flow data generated by each object in the period of time in the future.

[0069] Exemplarily, the first future cash flow of the first object and the second object in the next week has been predicted, and then the cash flow data between the known objects in the historical market cash flow data is used to simulate the cash flow data generated by the interaction between the first object, the second object and the reference object in the next week, and the cash flow data generated by the interaction between the first object, the second object and the reference object in the next week is used as the simulation result.

[0070] Optionally, the simulation of the interaction between various objects in the target simulated market is achieved by treating each object as a separate agent, and then setting a memory model for each agent so that it can store and recall past events. Exemplarily, taking object A as an example, the memory model of object A stores the flow data between object A and other objects. In the process of simulating the interaction between various objects, each agent will refer to the historical flow data in its memory model to simulate transaction behavior. In this way, multiple corporate agents conduct transaction interactions in the simulated market and generate future flow data. Each agent retrieves information from its own memory model to generate more realistic interaction behavior.

[0071] Among them, each agent refers to the historical flow data in its memory model, and the process of simulating trading behavior can be generated using a deep learning model or a rule engine.

[0072] Optionally, based on the historical flow data, the first future flow data between the first object and the second object is predicted, the reference object that interacts with the first object and the second object in the historical market flow data is queried, and based on the first future flow data and the historical market flow data, the second future flow data between each object in the target simulation market is simulated, and the data simulation results can be generated by the same deep learning model or rule engine.

[0073] In the above data simulation method, by receiving the historical flow data between the first object and the second object, and then predicting the first future flow data based on the historical flow data, past transaction patterns and behaviors can be fully understood. Secondly, by querying the reference objects that interact with the first object and the second object in the historical market flow data, other objects that have a significant impact on market transactions can be identified, thereby simulating the market environment more comprehensively. Finally, by combining the first future flow data and the historical market flow data, the second future flow data between the various objects in the target simulation market is simulated, and the transaction behavior between the various objects in the future time period is simulated, which can reflect the dynamic interaction between the various objects. In this way, not only the prediction accuracy is enhanced, but also the complex interactive relationships in the market are taken into account, and the market changes can be dynamically reflected, making the generated second future flow data more accurate.

[0074] In order to enhance the accuracy and reliability of market simulation, simulation is carried out by refining the time period.

[0075] Further, in one of the embodiments, the first future flow data and the second future flow data include first sub-flow data and second sub-flow data corresponding to multiple sub-time periods; based on the first future flow data and the historical market flow data, the second future flow data between each object in the target simulation market is simulated to obtain a data simulation result, including: based on the first sub-flow data corresponding to different sub-time periods and the historical market flow data, the second sub-flow data between each object in the target simulation market corresponding to different sub-time periods is simulated to obtain a data simulation result.

[0076] In this embodiment, the first future pipeline data is divided into a plurality of sub-time periods, and simulation is performed through the plurality of sub-time periods to obtain a plurality of second sub-pipeline data corresponding to the plurality of sub-time periods.

[0077] Optionally, in different sub-time periods, second sub-flow data between various objects in the target simulation market are simulated according to the first sub-flow data and historical market flow data to obtain simulation data results.

[0078] For example, assuming that the first sub-flow data is the flow data of object A and object B in the next week, the first sub-flow data is divided into 7 first sub-flow data in units of days. Then, 7 second sub-flow data are generated by simulating the first sub-flow data and the historical market flow data.

[0079] In the above embodiment, by refining the time period, the trend and pattern of transaction behavior in different time periods can be captured, thereby improving the accuracy of the overall prediction.

[0080] Furthermore, in one of the embodiments, based on the first sub-flow data corresponding to different sub-time periods and the historical market flow data, the second sub-flow data between each object in the target simulation market corresponding to the different sub-time periods is simulated to obtain a data simulation result, including: based on the first sub-flow data corresponding to the current sub-time period and the historical market flow data corresponding to the previous sub-time period, the second sub-flow data between each object in the target simulation market corresponding to the current sub-time period is simulated; based on the second sub-flow data, the historical market flow data is updated, and based on the updated historical market flow data and the first sub-flow data corresponding to the next sub-time period, the step of simulating the second sub-flow data corresponding to the next sub-time period is continued until the data simulation result is obtained.

[0081] In this embodiment, the second sub-flow data generated in each sub-time period is merged with the historical market flow data and used to predict the second sub-flow data corresponding to the next sub-time period.

[0082] Optionally, the first sub-flow data are sorted according to the time series, and then the first sub-flow data corresponding to the current sub-period are simulated with the historical market flow data corresponding to the previous sub-period in turn to obtain the second sub-flow data corresponding to different sub-periods. After the second sub-flow data corresponding to each sub-period is obtained, it will be updated with the historical market flow, and the updated historical flow data will be used to simulate the second sub-flow data corresponding to the next sub-period.

[0083] Exemplarily, the first future flow data is as follows:

[0084] Sub-time period date Sender Receiver Amount Way Sub-period 1 2024-08-03 Object A Object B 5000 purchase Sub-period 2 2024-08-04 Object B Object A 3000 Sale

[0085] Then, the first sub-flow data of sub-period 1 and the historical market flow data are used for simulation to obtain the second sub-flow data between the objects in the simulated target market in sub-period 1. Then, the second sub-flow data is merged with the historical market flow data as new historical market flow data. When simulating the second sub-flow data corresponding to sub-period 2, sub-period 2 and the new historical market flow data are used for simulation. Finally, multiple second sub-flow data are merged to obtain the second future flow data.

[0086] In the above embodiment, the market flow data can be gradually generated and updated in each sub-time period. This method can simulate market dynamics more accurately and provide more reliable simulation results by refining the time period and dynamically updating the historical market flow data.

[0087] However, in the process of continuously generating transaction data, there may be a situation where the context upper limit cannot be supported. At this time, it is necessary to extract important historical transaction data, that is, the target market transaction data mentioned in this embodiment.

[0088] Furthermore, in one of the embodiments, the historical market flow data is updated based on the second sub-flow data, including: integrating the second sub-flow data with the historical market flow data to obtain intermediate reference flow data; extracting target market flow data from the intermediate reference flow data, and using the target market flow data as updated historical market flow data.

[0089] The intermediate reference flow data refers to the flow data obtained by integrating the second sub-flow data with the historical market flow data.

[0090] In this embodiment, the second sub-flow data generated in each sub-time period is merged with the historical market flow data to obtain merged intermediate reference flow data. Then, the target market flow data is extracted from the intermediate reference flow data as updated historical market flow data.

[0091] Thus, in this embodiment, through the steps of integration, screening and updating, it is ensured that the historical market flow data can accurately reflect the key market dynamics. Updating the important flow data into the historical data helps to improve the accuracy of the market simulation.

[0092] Furthermore, in one of the embodiments, target market flow data is extracted from the intermediate reference flow data, including: calculating the recency index, importance index and relevance index of each flow data in the historical market flow data; calculating the comprehensive score of each flow data by combining the recency index, importance index and relevance index; and extracting from the intermediate reference flow data according to the comprehensive score to obtain the target market flow data.

[0093] In this embodiment, it is important to calculate the comprehensive score of each piece of transaction data based on recency, importance and relevance, and extract from the intermediate reference transaction data based on the comprehensive score to obtain the target market transaction data.

[0094] Among them, the recency index is used to measure the time proximity of the transaction data, and the higher the value, the newer the record; the importance index is used to measure the market importance of the transaction data, which is usually related to the transaction amount or transaction frequency; the relevance index is used to measure the relevance of the transaction data to the target market object. For example, the matching degree of the transaction object.

[0095] Optionally, the composite score is calculated as follows:

[0096] Score = α 1 ·recency+α 2 ·importance+α 3 ·relevance

[0097] Among them, recency is the recentness index, importance is the importance index, and relevance is the relevance index. 1 , α 2 and α 3 The specific weights can be set according to actual conditions and are not specifically limited in this embodiment.

[0098] Optionally, the time difference between the date of each running data and the current date can be calculated as the recency index. For example, the recency index = current date - running data date.

[0099] Optionally, the importance index may be calculated by calculating the ratio between the transaction amount of each transaction data and the maximum transaction amount. For example, importance index = transaction amount / maximum transaction amount.

[0100] Optionally, the ratio between the transaction frequency with the target object and the total transaction frequency can be calculated as the correlation index. For example, correlation index = number of transactions with the target object / total number of transactions.

[0101] In the above embodiment, by calculating the recency index, importance index and relevance index of each flow data and combining these indexes to obtain a comprehensive score, important flow data can be extracted.

[0102] In one of the embodiments, based on the first future flow data and the historical market flow data, the second future flow data between the objects in the target simulation market is simulated to obtain the data simulation result, which also includes: obtaining simulation parameters; the simulation parameters include one or more of the enterprise operation strategy, the competitor strategy, the external environment parameters, and the object attributes; based on the simulation parameters, the first future flow data and the historical market flow data, the second future flow data between the first object and the second object and the reference object is simulated to obtain the data simulation result.

[0103] In addition to simulating the second future flow data between various objects in the target simulation market based on the first sub-flow data and the historical market flow data as mentioned in the above embodiment, in the present embodiment, simulation parameters are also obtained and simulation is performed based on the simulation parameters, so that the simulation results are more in line with the actual mining.

[0104] That is to say, in this embodiment, a simulated market environment is constructed by acquiring simulation parameters, and various objects are made to interact in the simulated market environment.

[0105] The simulation parameters may include one or more of the enterprise operation strategy, competitor strategy, external environment parameters, and object attributes. The object attribute refers to the character of the object. For example, the object attribute may be a big spender, cautious, financing, or going bankrupt. The specific simulation process may refer to the records in the above embodiments, and will not be described repeatedly here.

[0106] In this embodiment, the simulation parameters, the first sub-flow data and the historical market flow data are combined to accurately simulate the interaction between the objects in the target market in the future time period. This method not only takes into account the historical data, but also integrates the strategies and environmental factors in reality, providing a more comprehensive and dynamic market simulation result.

[0107] In one of the embodiments, based on historical flow data, the first sub-flow data between the first object and the second object is predicted by a pre-trained flow prediction model; the prediction process of the flow prediction model includes: obtaining sample flow data; the sample flow data carries a flow label; inputting the sample flow data into the initial model for training to obtain a predicted flow; obtaining a target loss function based on the predicted flow and the flow label, and adjusting the parameters of the initial model according to the target loss function until the training is completed to obtain a flow prediction model.

[0108] The sample transaction data refers to the sample data used to train the transaction prediction model, which includes information such as the sender, receiver, amount, time, and behavior. It should be noted that the sample data needs to cover various transaction records to ensure the generalization ability of the model; the transaction label refers to the actual target value of each record.

[0109] The initial model refers to a machine learning model that has not been trained and whose parameters are still in the initialization stage. For example, the initial model can be a Transformer model.

[0110] Optionally, the sample flow data is input into the initial model for training to generate predicted flow data. The target loss function is calculated based on the predicted flow data and flow labels. The optimization algorithm is used to extract the model parameters of the second ring to minimize the loss function. The process of forward propagation, loss calculation and parameter optimization is repeated until the model training is completed and the flow prediction model is obtained.

[0111] In the above embodiment, the model is continuously iterated to train and optimize so that it can accurately predict future water flow data based on historical data. Sample data and actual labels are used during model training to optimize the target loss function and ensure that the model can effectively perform water flow prediction.

[0112] In one of the embodiments, before the sample flow data is input into the initial model for training, the process also includes: preprocessing the sample flow data; the preprocessing includes one or more of anonymization and cleaning.

[0113] Before inputting the sample pipeline into the initial model for training, the sample pipeline data needs to be preprocessed.

[0114] The preprocessing includes one or more of anonymization and cleaning.

[0115] Among them, anonymization is to remove identity information to protect the privacy of the subject.

[0116] Among them, cleaning refers to processing missing values ​​and outliers on sample flow data. In other embodiments, preprocessing may also include data format conversion, feature normalization, feature engineering, etc., which are specifically explained in detail here.

[0117] In the above embodiment, the preprocessing step can effectively improve the prediction ability and robustness of the model, ensuring the accuracy and reliability of flow prediction.

[0118] In one of the embodiments, the method further includes: obtaining an objective function and constraints; and calculating a target decision plan corresponding to the data simulation result based on the objective function and constraints.

[0119] First, the objective function and constraints of the optimization problem need to be clearly defined.

[0120] Assume that the goal is to maximize corporate profits and that the company has budget and resource constraints in the production and sales process.

[0121] For example, the objective function can be

[0122]

[0123] Among them, Pi is the selling price of the i-th product, Ci is the cost of the i-th product, and xi is the output of the i-th product.

[0124] Illustratively, the constraints may be budget constraints and resource constraints.

[0125] The budget limit is

[0126]

[0127] The resource limit is

[0128]

[0129] Among them, R ij is the amount of the jth resource required for the ith product.

[0130] According to the above objective function and constraint conditions, a linear programming model is established, and the linear programming model is solved to obtain the target parameters. In this embodiment, the target parameters are the optimal production volume of each product and the maximum profit that the enterprise can achieve under these production volumes.

[0131] In the above embodiment, an optimization algorithm combining linear programming and simulation engine can be used to make the best business decision in a complex and changing market environment.

[0132] In an exemplary embodiment, in combination Figure 2 As shown, Figure 2 A schematic diagram of data simulation in one embodiment.

[0133] Figure 2 The simulation engine is used to receive user input, that is, the historical flow data between the first object and the second object and one or more of the simulation parameters, and then execute prediction, simulation and other processes. For details, please refer to the records in the above embodiments and will not be described in detail again.

[0134] In this embodiment, data visualization can also be achieved. In this embodiment, a multi-dimensional data visualization method is used to present a visualization display of cash flow simulation records under multiple strategies of the research enterprise, including cash flow trends, transaction network diagrams, and the impact of competitor strategies.

[0135] Finally, a simulation report can be output based on the simulation results. The simulation report includes simulated cash flow records, strategy evaluation, optimization suggestions and visual analysis results.

[0136] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0137] Based on the same inventive concept, the embodiment of the present application also provides a data simulation device for implementing the data simulation method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more data simulation device embodiments provided below can refer to the limitations on the data simulation method above, and will not be repeated here.

[0138] In an exemplary embodiment, Figure 3 As shown, a data simulation device is provided, including: a receiving module 310, a prediction module 320, a query module 330 and a simulation module 340, wherein:

[0139] The receiving module 310 is used to receive the historical flow data between the first object and the second object.

[0140] The prediction module 320 is used to predict first future flow data between the first object and the second object based on the historical flow data.

[0141] The query module 330 is used to query the reference objects that interact with the first object and the second object in the historical market flow data, and obtain the reference objects.

[0142] The simulation module 340 is used to simulate the second future flow data between various objects in the target simulation market according to the first future flow data and the historical market flow data to obtain data simulation results; the target simulation market includes the first object, the second object and the reference object.

[0143] In one embodiment, the simulation module 340 includes:

[0144] The time period simulation unit is used to simulate the second sub-flow data between various objects in the target simulation market corresponding to different sub-time periods according to the first sub-flow data corresponding to different sub-time periods and the historical market flow data to obtain data simulation results.

[0145] In one embodiment, the time period simulation unit includes:

[0146] The first simulation subunit is used to simulate the second sub-flow data between various objects in the target simulation market corresponding to the current sub-time period according to the first sub-flow data corresponding to the current sub-time period and the historical market flow data corresponding to the previous sub-time period.

[0147] The second simulation subunit is used to simulate the second sub-flow data between various objects in the target simulation market corresponding to the current sub-time period according to the first sub-flow data corresponding to the current sub-time period and the historical market flow data corresponding to the previous sub-time period.

[0148] In one embodiment, the second simulation subunit further includes:

[0149] The integration subunit is used to integrate the second sub-flow data with the historical market flow data to obtain intermediate reference flow data.

[0150] The extraction subunit is used to extract the target market flow data from the intermediate reference flow data, and use the target market flow data as the updated historical market flow data.

[0151] In one embodiment, the extraction subunit includes:

[0152] The indicator calculation subunit is used to calculate the recency indicator, importance indicator and correlation indicator of each flow data in the historical market flow data.

[0153] The score calculation subunit is used to calculate the comprehensive score of each flow data by comprehensively considering the recency index, importance index and relevance index.

[0154] The extraction subunit is used to extract from the intermediate reference flow data according to the comprehensive score to obtain the target market flow data.

[0155] In one embodiment, the simulation module 340 further includes

[0156] The parameter acquisition unit is used to acquire simulation parameters; the simulation parameters include one or more of enterprise operation strategy, competitor strategy, external environment parameters, and object attributes.

[0157] The parameter simulation unit is used to simulate the first object and the second future flow data between the second object and the reference object according to the simulation parameters, the first sub-flow data and the historical market flow data to obtain the data simulation result.

[0158] In one embodiment, the prediction module includes:

[0159] The sample data acquisition unit is used to acquire sample flow data; the sample flow data carries a flow tag.

[0160] The flow prediction unit is used to input sample flow data into the initial model for training to obtain the predicted flow.

[0161] The model training unit is used to obtain the target loss function based on the predicted flow and flow label, and adjust the parameters of the initial model according to the target loss function until the training is completed to obtain the flow prediction model.

[0162] In one embodiment, the prediction module further includes:

[0163] The preprocessing unit is used to preprocess the sample flow data; the preprocessing includes one or more of anonymization and cleaning.

[0164] In one embodiment, the above device further includes an optimization module, and the optimization module includes:

[0165] The condition acquisition unit is used to obtain the objective function and constraint conditions.

[0166] The optimization unit is used to calculate the target decision plan corresponding to the data simulation results based on the objective function and constraints.

[0167] Each module in the above data simulation device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0168] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 4As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store historical market flow data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a data simulation method is implemented.

[0169] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0170] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the following steps when executing the computer program: receiving historical flow data between a first object and a second object; predicting first future flow data between the first object and the second object based on the historical flow data; querying reference objects that interact with the first object and the second object in the historical market flow data; simulating second future flow data between objects in a target simulation market based on the first future flow data and the historical market flow data to obtain data simulation results; the target simulation market includes the first object, the second object and the reference object.

[0171] In one embodiment, when the processor executes the computer program, the following steps are also implemented: based on the first sub-flow data corresponding to different sub-time periods and the historical market flow data, the second sub-flow data between the objects in the target simulation market corresponding to the different sub-time periods are simulated to obtain data simulation results.

[0172] In one embodiment, when the processor executes the computer program, the following steps are also implemented: based on the first sub-flow data corresponding to the current sub-time period and the historical market flow data corresponding to the previous sub-time period, the second sub-flow data between the objects in the target simulation market corresponding to the current sub-time period is simulated; based on the second sub-flow data, the historical market flow data is updated, and based on the updated historical market flow data and the first sub-flow data corresponding to the next sub-time period, the step of simulating the second sub-flow data corresponding to the next sub-time period is continued until the data simulation result is obtained.

[0173] In one embodiment, when the processor executes the computer program, the following steps are also implemented: integrating the second sub-flow data with the historical market flow data to obtain intermediate reference flow data; extracting the target market flow data from the intermediate reference flow data, and using the target market flow data as the updated historical market flow data.

[0174] In one embodiment, when the processor executes the computer program, the following steps are also implemented: calculating the recency index, importance index and relevance index of each flow data in the historical market flow data; calculating the comprehensive score of each flow data by combining the recency index, importance index and relevance index; extracting from the intermediate reference flow data according to the comprehensive score to obtain the target market flow data.

[0175] In one embodiment, when the processor executes the computer program, it also implements the following steps: obtaining simulation parameters; the simulation parameters include one or more of enterprise operating strategies, competitor strategies, external environment parameters, and object attributes; based on the simulation parameters, the first sub-flow data, and the historical market flow data, simulating the first object and the second future flow data between the second object and the reference object to obtain data simulation results.

[0176] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining sample flow data; the sample flow data carries a flow label; inputting the sample flow data into the initial model for training to obtain a predicted flow; obtaining a target loss function based on the predicted flow and the flow label, and adjusting the parameters of the initial model according to the target loss function until the training is completed to obtain a flow prediction model.

[0177] In one embodiment, when the processor executes the computer program, the following steps are also implemented: preprocessing the sample flow data; the preprocessing includes one or more of anonymization and cleaning.

[0178] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining the objective function and the constraints; and calculating the target decision plan corresponding to the data simulation result based on the objective function and the constraints.

[0179] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented: receiving historical flow data between a first object and a second object; predicting first future flow data between the first object and the second object based on the historical flow data; querying reference objects that interact with the first object and the second object in the historical market flow data; simulating second future flow data between objects in a target simulation market based on the first future flow data and the historical market flow data to obtain data simulation results; the target simulation market includes the first object, the second object and the reference object.

[0180] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: based on the first sub-flow data corresponding to different sub-time periods and the historical market flow data, the second sub-flow data between the objects in the target simulation market corresponding to the different sub-time periods are simulated to obtain data simulation results.

[0181] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: based on the first sub-flow data corresponding to the current sub-time period and the historical market flow data corresponding to the previous sub-time period, the second sub-flow data between the objects in the target simulation market corresponding to the current sub-time period is simulated; based on the second sub-flow data, the historical market flow data is updated, and based on the updated historical market flow data and the first sub-flow data corresponding to the next sub-time period, the step of simulating the second sub-flow data corresponding to the next sub-time period is continued until the data simulation result is obtained.

[0182] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: integrating the second sub-flow data with the historical market flow data to obtain intermediate reference flow data; extracting the target market flow data from the intermediate reference flow data, and using the target market flow data as the updated historical market flow data.

[0183] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: calculating the recency index, importance index and relevance index of each transaction data in the historical market transaction data; calculating the comprehensive score of each transaction data by combining the recency index, importance index and relevance index; extracting from the intermediate reference transaction data according to the comprehensive score to obtain the target market transaction data.

[0184] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining simulation parameters; the simulation parameters include one or more of enterprise operating strategies, competitor strategies, external environment parameters, and object attributes; based on the simulation parameters, the first sub-flow data, and the historical market flow data, simulating the first object and the second future flow data between the second object and the reference object to obtain data simulation results.

[0185] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining sample flow data; the sample flow data carries a flow label; inputting the sample flow data into the initial model for training to obtain a predicted flow; obtaining a target loss function based on the predicted flow and the flow label, and adjusting the parameters of the initial model according to the target loss function until the training is completed to obtain a flow prediction model.

[0186] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: preprocessing the sample flow data; the preprocessing includes one or more of anonymization and cleaning.

[0187] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented: obtaining the objective function and the constraints; and calculating the target decision plan corresponding to the data simulation result based on the objective function and the constraints.

[0188] In one embodiment, a computer program product is provided, including a computer program, which implements the following steps when executed by a processor: receiving historical flow data between a first object and a second object; predicting first future flow data between the first object and the second object based on the historical flow data; querying reference objects that interact with the first object and the second object in the historical market flow data; simulating second future flow data between objects in a target simulation market based on the first future flow data and the historical market flow data to obtain data simulation results; the target simulation market includes the first object, the second object and the reference object.

[0189] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: based on the first sub-flow data corresponding to different sub-time periods and the historical market flow data, the second sub-flow data between the objects in the target simulation market corresponding to the different sub-time periods are simulated to obtain data simulation results.

[0190] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: based on the first sub-flow data corresponding to the current sub-time period and the historical market flow data corresponding to the previous sub-time period, the second sub-flow data between the objects in the target simulation market corresponding to the current sub-time period is simulated; based on the second sub-flow data, the historical market flow data is updated, and based on the updated historical market flow data and the first sub-flow data corresponding to the next sub-time period, the step of simulating the second sub-flow data corresponding to the next sub-time period is continued until the data simulation result is obtained.

[0191] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: integrating the second sub-flow data with the historical market flow data to obtain intermediate reference flow data; extracting the target market flow data from the intermediate reference flow data, and using the target market flow data as the updated historical market flow data.

[0192] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: calculating the recency index, importance index and relevance index of each transaction data in the historical market transaction data; calculating the comprehensive score of each transaction data by combining the recency index, importance index and relevance index; extracting from the intermediate reference transaction data according to the comprehensive score to obtain the target market transaction data.

[0193] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining simulation parameters; the simulation parameters include one or more of enterprise operating strategies, competitor strategies, external environment parameters, and object attributes; based on the simulation parameters, the first sub-flow data, and the historical market flow data, simulating the first object and the second future flow data between the second object and the reference object to obtain data simulation results.

[0194] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining sample flow data; the sample flow data carries a flow label; inputting the sample flow data into the initial model for training to obtain a predicted flow; obtaining a target loss function based on the predicted flow and the flow label, and adjusting the parameters of the initial model according to the target loss function until the training is completed to obtain a flow prediction model.

[0195] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: preprocessing the sample flow data; the preprocessing includes one or more of anonymization and cleaning.

[0196] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented: obtaining the objective function and the constraints; and calculating the target decision plan corresponding to the data simulation result based on the objective function and the constraints.

[0197] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0198] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0199] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A data simulation method, characterized in that: The method comprises: Receiving historical flow data between the first object and the second object; Predicting first future pipeline data between the first object and the second object based on the historical pipeline data; Querying reference objects that interact with the first object and the second object in historical market flow data; According to the first future flow data and the historical market flow data, the second future flow data between the objects in the target simulation market is simulated to obtain a data simulation result; the target simulation market includes the first object, the second object and the reference object.

2. The data simulation method according to claim 1, characterized in that: The first future pipeline data and the second future pipeline data include first sub-pipeline data and second sub-pipeline data corresponding to a plurality of sub-time periods; The step of simulating the second future flow data between the objects in the target simulation market according to the first sub-flow data and the historical market flow data to obtain a data simulation result includes: According to the first sub-flow data and historical market flow data corresponding to different sub-time periods, the second sub-flow data between various objects in the target simulation market corresponding to different sub-time periods are simulated to obtain the data simulation result.

3. The data simulation method according to claim 2, characterized in that: The step of simulating the second sub-flow data between the objects in the target simulation market corresponding to the different sub-time periods according to the first sub-flow data and the historical market flow data corresponding to the different sub-time periods to obtain the data simulation result includes: According to the first sub-flow data corresponding to the current sub-time period and the historical market flow data corresponding to the previous sub-time period, simulating the second sub-flow data between various objects in the target simulation market corresponding to the current sub-time period; The historical market flow data is updated based on the second sub-flow data, and according to the updated historical market flow data and the first sub-flow data corresponding to the next sub-time period, the step of simulating the second sub-flow data corresponding to the next sub-time period is continued until the data simulation result is obtained; The updating of the historical market transaction data based on the second sub-transaction data includes: Integrate the second sub-flow data with the historical market flow data to obtain intermediate reference flow data; Extracting target market flow data from the intermediate reference flow data, and using the target market flow data as the updated historical market flow data; The step of extracting target market flow data from the intermediate reference flow data includes: Calculating the recency index, importance index and relevance index of each transaction data in the historical market transaction data; Calculate a comprehensive score for each of the flow data by comprehensively considering the recency index, the importance index and the relevance index; The target market flow data is extracted from the intermediate reference flow data according to the comprehensive score to obtain the target market flow data.

4. The data simulation method according to claim 1, characterized in that: The step of simulating second future flow data between various objects in the target simulation market according to the first sub-flow data and the historical market flow data to obtain a data simulation result further includes: Acquiring simulation parameters; the simulation parameters include one or more of enterprise operation strategy, competitor strategy, external environment parameters, and object attributes; According to the simulation parameters, the first sub-flow data and the historical market flow data, the first object and the second future flow data between the second object and the reference object are simulated to obtain the data simulation result.

5. The data simulation method according to claim 1, characterized in that: The first sub-flow data between the first object and the second object is predicted based on the historical flow data by a pre-trained flow prediction model; the prediction process of the flow prediction model includes: Acquire sample flow data; the sample flow data carries a flow tag; Inputting the sample flow data into the initial model for training to obtain the predicted flow; Obtaining a target loss function based on the predicted flow and the flow label, and adjusting the parameters of the initial model according to the target loss function until the training is completed to obtain the flow prediction model; Before the sample pipeline data is input into the initial model for training, the following steps are also included: Preprocessing the sample flow data; The preprocessing includes one or more of anonymization and cleaning.

6. The data simulation method according to any one of claims 1 to 5, characterized in that: The method further comprises: Get the objective function and constraints; Based on the objective function and the constraints, a target decision plan corresponding to the data simulation result is calculated.

7. A data simulation device, characterized in that: The device comprises: A receiving module, used for receiving historical flow data between the first object and the second object; A prediction module, configured to predict first future pipeline data between the first object and the second object based on the historical pipeline data; A query module, used to query reference objects that interact with the first object and the second object in historical market flow data, and obtain the reference objects; A simulation module is used to simulate the second future flow data between various objects in the target simulation market according to the first future flow data and the historical market flow data to obtain data simulation results; the target simulation market includes the first object, the second object and the reference object.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.