Digital twinning-based monopoly situation prediction method and device

Abstract

The invention discloses a monopoly situation prediction method and device based on digital twinning, and the method comprises the steps: building a monopoly behavior data target range of a digital space based on a digital twinning technology in combination with an intelligent algorithm model, carrying out the recognition and evaluation of a monopoly behavior in the competition of a market subject through the combination of simulation data and real data, and carrying out the prediction of the monopoly situation based on digital twinning. And the monopoly situation can be effectively predicted, and early warning of monopoly illegal behaviors is realized.

Description

Technical Field

[0001] The present invention relates to the field of digital twin technology, and in particular to a method and device for predicting a monopoly situation based on digital twins. Background Art

[0002] In the new economic era, as the speed of transformation of new technologies into the market has significantly accelerated, and global marketing and information networks have been formed, the speed at which new products enter the market and break the original market monopoly has greatly accelerated. Generally speaking, market monopoly status can be proved by a company's large share of the relevant market, or by indirect evidence such as price control or the ability to exclude competition. However, it usually relies on manual data sorting and evidence collection, which is limited by objective factors such as manpower and material resources, and the judgment efficiency and result accuracy are low. The existing technology has not formed an algorithm model suitable for the antitrust field, and it is impossible to form quantitative analysis and prediction results based on actual data. The level of intelligence is poor and cannot adapt to the rapid development of the new economic era. Summary of the invention

[0003] In response to the technical problems existing in the prior art, the present invention provides a monopoly situation prediction method based on digital twins. Based on digital twin technology and combined with an intelligent algorithm model, it can identify and evaluate the monopolistic behaviors in the competition among market players and effectively predict the monopoly situation.

[0004] In a first aspect, an embodiment of the present invention provides a method for predicting a monopoly situation based on digital twins, the method comprising:

[0005] Step 1: Obtain real data on market competition behaviors in multiple industries and platforms, extract key features using text mining data, create a monopoly behavior feature library and a monopoly behavior evaluation index system, and set weight values ​​for each indicator in the evaluation index system;

[0006] Step 2: Based on the monopolistic behavior feature library and the monopolistic behavior evaluation index system constructed in step 1, generate market competition behavior simulation data, and input them into the generative adversarial network, adjust the parameters of the simulation data, obtain the simulation data that conforms to the distribution of monopolistic behavior characteristics, and construct a monopolistic behavior data target range in the digital space;

[0007] Step 3: Use the data-driven modeling method to build a digital twin prediction model, simulate market competition behavior in the monopoly behavior data target range, input the simulated data into the digital twin prediction model to obtain the monopoly situation prediction result, and use the real data for external verification, and adjust the operating parameters of the digital twin prediction model based on the verification result;

[0008] Step 4: Obtain the target economic entity information to be monitored and the target economic environment information of the industry in which it is located, pre-process the target economic entity information and the target economic environment information, use the digital twin prediction model to perform dynamic monitoring, and output the monopoly situation assessment results, and send the assessment results to the supervision platform through the communication module.

[0009] In a possible implementation of the present invention, the method provided by the embodiment of the present invention further includes:

[0010] Setting weight values ​​for each indicator in the evaluation indicator system specifically includes:

[0011] The expert scoring method is used to set initial weights for each evaluation indicator in the evaluation indicator system, and then the entropy weighting method is used to adjust and optimize the initial weights to obtain the final weight values ​​of each evaluation indicator.

[0012] In a possible implementation of the present invention, the method provided by the embodiment of the present invention further includes:

[0013] The monopolistic behavior data target range in the digital space includes a core data warehouse part, a simulation data generation part, and a digital twin visualization part. The core data warehouse is used to collect, analyze and clean up the original data, and generate the original data layer, data integration layer and feature data layer of the core data warehouse; the simulation data generation part is used to learn data features and model according to the features; the digital twin visualization part is used to visualize the features of the data and the effects of the model.

[0014] In a possible implementation of the present invention, the method provided by the embodiment of the present invention further includes:

[0015] Preprocessing the target economic entity information and the target economic environment information specifically includes: after denoising and cleaning the target economic entity information and the target economic environment information, extracting parameters corresponding to key features based on the monopoly behavior feature library, and performing quantization processing to obtain feature vectors.

[0016] In a possible implementation of the present invention, the method provided by the embodiment of the present invention further includes:

[0017] The regulatory platform includes one or more of the following: the regulatory platform of the target economic entity, the regulatory platform of the market regulatory department, and the third-party regulatory platform.

[0018] In the second aspect, an embodiment of the present invention provides a monopoly situation prediction device based on digital twins, including a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the above-mentioned monopoly situation prediction method based on digital twins.

[0019] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0020] 1. The present invention makes antitrust law enforcement more professional and refined, greatly improving the credibility and effectiveness of antitrust work.

[0021] 2. The present invention strengthens and innovates market price supervision.

[0022] 3. The present invention helps to timely discover problems in the business operation process of an enterprise and effectively reduces the cost of correcting violations.

[0023] 4. The present invention ensures the efficiency of data utilization, verifies the effectiveness of antitrust detection work, enriches the monitoring dimensions, and effectively improves the reliability of antitrust monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0025] Figure 1 It is a method flow chart of a method for predicting monopoly situation based on digital twins according to an exemplary embodiment.

[0026] Figure 2 It is a diagram showing a data range structure of monopolistic behavior in a digital space according to an exemplary embodiment. DETAILED DESCRIPTION

[0027] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be comprehensive and complete and will fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted.

[0028] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the present disclosure.

[0029] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0030] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0031] Embodiment 1

[0032] Figure 1 The method flow chart of a method for predicting a monopoly situation based on digital twins is shown according to an exemplary embodiment. The method for predicting a monopoly situation based on digital twins at least includes steps S1 to S4.

[0033] Step S1, obtain real data on market competition behaviors in multiple industries and platforms, use text mining data to extract key features, create a monopoly behavior feature library and a monopoly behavior evaluation index system, and set corresponding weights for each indicator in the evaluation index system.

[0034] Market competition behavior mainly refers to suspected monopolistic behaviors such as monopoly agreements, concentration of operators not reported in accordance with the law, unfair prices, sales below cost, differential treatment, etc., which can be obtained from regulatory authorities or public historical monopoly cases. Information on economic entities suspected of monopolistic behavior and information on the economic environment of the industry in which they are located can be obtained. Information on economic entities and economic environment refers to relevant information used to determine whether the business behavior of economic entities is suspected of monopolistic behavior, such as basic information data of enterprises, tax data, sales data, market share, as well as market price discrimination, market concentration, degree of product differentiation, price fluctuations, industry growth rate and consumer demand, etc. Text mining technology is used to extract key features from them, and a monopolistic behavior feature library is constructed. The obtained real data of market competition behavior is used to verify the monopolistic behavior feature library, and the validity and reliability of the monopolistic behavior feature library are verified according to the degree of matching between the real data of market competition behavior and the monopolistic behavior feature library. The monopolistic behavior feature library is adjusted based on the verification results to finally obtain a true and reliable monopolistic behavior feature library.

[0035] Based on the existing rules for judging monopolistic behavior, a monopolistic behavior evaluation index system is constructed. By way of example, the present application provides a method for establishing a monopolistic behavior evaluation index system. For example, a three-level evaluation index can be constructed, with the monopolization of factual standards as the target layer of the index system, the five levels of technology, market, management, policy and law as the criterion layer of the index system, and the issues corresponding to the five levels as the element layer of the index system, as shown in Table 1.

[0036] Those skilled in the art should know that, in addition to the above-mentioned methods, other methods may be used to establish an evaluation index system according to actual needs. The established evaluation index system may be a two-level system, or a three-level or four-level system, and the present invention is not limited to this.

[0037] Table 1 Monopoly behavior evaluation index system

[0038]

[0039] Based on expert experience, the expert scoring method is used to set the initial weights for each indicator in the evaluation index system, and then the entropy weighting method is used to adjust and optimize the initial weights to obtain the final weight values ​​of each evaluation indicator.

[0040] Step S2, based on the monopolistic behavior feature library and monopolistic behavior evaluation index system constructed in step S1, generates market competition behavior simulation data, and inputs it into the generative adversarial network, adjusts the parameters of the simulation data, obtains simulation data that conforms to the distribution of monopolistic behavior characteristics, and constructs a monopolistic behavior data target range in the digital space.

[0041] The generative adversarial network consists of a generator G and a discriminator D. During the entire training process, G and D are the two "game" parties. The generator G captures the distribution of sample data, and the discriminator D is a binary classifier, which is used to judge the probability that the input result comes from the training data. Both G and D are nonlinear mapping functions, which are multi-layer perceptrons or convolutional neural networks. During the training process, the goal of the generator G is to generate results that are as close to the original data as possible to deceive the discriminator D; and the goal of D is to distinguish the results generated by G from the real data as much as possible. G and D form a dynamic "game process", and finally obtain simulated data whose feature distribution is extremely similar to the real data, which serves as the data base for the data target range of monopoly behavior.

[0042] like Figure 2 As shown in the figure, the data range of monopoly behavior in digital space can be divided into the core data warehouse part, the simulation data generation part, and the digital twin visualization part according to data, models, and applications. The core data warehouse part aims to collect, analyze, and clean up the original data to generate the original data layer of the core data warehouse. Then, the comparison and mapping of data from different data sources are completed to generate the data integration layer of the core data warehouse. Finally, the characteristics of the data are processed to generate the characteristic data layer of the core data warehouse.

[0043] The simulation data generation part is the process of learning data features and modeling based on the features. The first step is to generate a training data set that can be used for model training by processing the feature engineering of the feature data layer data. The second step is to select the appropriate model and parameters to perform feature learning and generate simulation data. The third step is to compare and evaluate the simulation data generated by the model, analyze the model performance, and optimize the model. The fourth step is to publish the trained model. The model training process requires repeated iterations and optimization. The results of model testing and evaluation provide a basis for the selection of different models, prompting the selection of continuously optimized training models.

[0044] The digital twin visualization part is the process of publishing the previous model capabilities and monopoly behavior data capabilities. Through the digital twin visualization module, the simulation data model can be used to generate monopoly behavior simulation data. The simulation data of monopoly behavior can also be stored in the feature data layer of the data warehouse to enrich the accumulation of data. Through the accumulated feature data, the monopoly behavior can be simulated in the visualization part, and the characteristics of the data and the effects of the model can be visualized.

[0045] Step S3, use the data-driven modeling method to build a digital twin prediction model, simulate the competitive behavior of market players in the data target range, input the simulated data into the digital twin prediction model to obtain the monopoly situation prediction results, and use real data for external verification, and adjust the operating parameters of the digital twin prediction model based on the verification results.

[0046] Digital twin refers to the integration of multidisciplinary and multi-scale simulation processes by making full use of data such as physical models, sensors, and operation history. As a mirror image of physical products in virtual space, it reflects the entire life cycle of the corresponding physical entity products. The essence of digital twin is to create a twin model of the physical entity and simulate it with the twin model as the basic model to reflect the real operating status of the physical entity in real time, and adjust the operating parameters of the physical entity through the feedback of the twin model to achieve the effect of optimization. The twin model has two significant characteristics: the twin model and the object it is to reflect are basically the same in appearance (geometric size and shape), content (structural composition and its macro / micro physical properties) and properties (function and performance); it allows the real operating status / state to be mirrored / reflected through simulation and other means.

[0047] The data-driven modeling method uses data mining technology to find useful information between data and establish a more specific and clear function expression to describe the relationship from input variables to output variables. It aims to fit samples and has a fixed input-output relationship. It can construct parameter optimization functions, select appropriate data-driven models and appropriate model structures from known data-driven models, and establish mathematical relationship expressions for corresponding models. Generally, BP neural network models, response surface models, support vector machines, etc. are often selected.

[0048] A digital twin prediction model is constructed using data-driven modeling methods, which treats market entities as physical entities, identifies their current status, and predicts their future status, effectively improving the reliability of monopoly behavior monitoring results.

[0049] Step S4, obtain the target economic entity information to be monitored and the target economic environment information of the industry in which it is located, pre-process the target economic entity information and the target economic environment information, use the digital twin prediction model to perform dynamic monitoring, and output the monopoly situation assessment results, and send the assessment results to the supervision platform through the communication module.

[0050] In practical applications, the target object to be detected is obtained, such as an important business entity in a key industry, and the information of the target economic entity and the target economic environment of the industry in which it is located is obtained. After denoising and cleaning the obtained information, the parameters corresponding to the key features are extracted based on the monopoly behavior feature library, and quantified to obtain the feature vector. Then, the digital twin prediction model is used to obtain the monopoly situation assessment results, and the assessment results are provided to the regulatory platform.

[0051] The regulatory platform includes the regulatory platform of the target economic entities, the regulatory platform of the market regulatory department and the third-party regulatory platform. The target economic entities can conduct self-inspection and rectification of their own business operations based on the evaluation results. The market regulatory department can intervene in advance in potential monopolistic behaviors in the market based on the evaluation results, and can also effectively supervise monopolistic behaviors that have already occurred.

[0052] In the process of using the digital twin prediction model to dynamically monitor the target economic entities, visualization parameters can be set to visualize the actual operation process of the digital space.

[0053] Embodiment 2

[0054] An embodiment of the present invention also provides a monopoly situation prediction device based on digital twins, including a processor, a memory, and a program or instruction stored in the memory and executable on the processor. When the program or instruction is executed by the processor, the above-mentioned monopoly situation prediction method based on digital twins is implemented.

[0055] It should be noted that, in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "including a ..." do not exclude the presence of other identical factors in the process, method, article or device including the elements.

[0056] A person of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, etc., various media that can store program codes.

[0057] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A monopoly situation prediction method based on digital twins, characterized in that: The following steps are involved: Step 1: Obtain real data on market competition behaviors in multiple industries and platforms, extract key features using text mining data, create a monopoly behavior feature library and a monopoly behavior evaluation index system, and set weight values ​​for each indicator in the evaluation index system; Step 2: Based on the monopolistic behavior feature library and the monopolistic behavior evaluation index system constructed in step 1, generate market competition behavior simulation data, and input them into the generative adversarial network, adjust the parameters of the simulation data, obtain the simulation data that conforms to the distribution of monopolistic behavior characteristics, and construct a monopolistic behavior data target range in the digital space; Step 3: Use the data-driven modeling method to build a digital twin prediction model, simulate market competition behavior in the monopoly behavior data target range, input the simulated data into the digital twin prediction model to obtain the monopoly situation prediction result, and use the real data for external verification, and adjust the operating parameters of the digital twin prediction model based on the verification result; Step 4: Obtain the target economic entity information to be monitored and the target economic environment information of the industry in which it is located, pre-process the target economic entity information and the target economic environment information, use the digital twin prediction model to perform dynamic monitoring, and output the monopoly situation assessment results, and send the assessment results to the supervision platform through the communication module.

2. The method according to claim 1, characterized in that: Setting weight values ​​for each indicator in the evaluation indicator system specifically includes: The expert scoring method is used to set initial weights for each evaluation indicator in the evaluation indicator system, and then the entropy weighting method is used to adjust and optimize the initial weights to obtain the final weight values ​​of each evaluation indicator.

3. The method according to claim 1, characterized in that The monopoly behavior data range in the digital space includes a core data warehouse part, a simulation data generation part, and a digital twin visualization part. The core data warehouse is used to collect, analyze and clean up the original data, and generate the original data layer, data integration layer and feature data layer of the core data warehouse; the simulation data generation part is used to learn data features and model according to the features; The digital twin visualization part is used to visualize the characteristics of the data and the effects of the model.

4. The method according to claim 3, characterized in that Preprocessing the target economic entity information and the target economic environment information specifically includes: after denoising and cleaning the target economic entity information and the target economic environment information, extracting parameters corresponding to key features based on the monopoly behavior feature library, and performing quantization processing to obtain feature vectors.

5. The method according to claim 4, characterized in that The regulatory platform includes one or more of the following: the regulatory platform of the target economic entity, the regulatory platform of the market regulatory department, and the third-party regulatory platform.

6. A monopoly situation prediction device based on digital twins, comprising a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein when the program or instruction is executed by the processor, the monopoly situation prediction method based on digital twins is implemented.

7. A prediction method based on digital twins, characterized in that: The following steps are involved: Step 1: Obtain data A, extract key features using text mining data, create a feature library and an evaluation index system, and set weight values ​​for each index in the evaluation index system; Step 2: Generate simulated data based on the feature library and the evaluation index system constructed in step 1, and input them into the generative adversarial network, adjust the parameters of the simulated data, and construct a data target range in the digital space; Step three: Use the data-driven modeling method to build a digital twin prediction model, input the simulation data into the digital twin prediction model to obtain the monopoly situation prediction result, and use the data A for external verification, and adjust the operating parameters of the digital twin prediction model based on the verification results.

8. The method according to claim 7, characterized in that The data A is real data on market competition behaviors in multiple industries and platforms; the feature library is a monopoly behavior feature library; the evaluation index system is a monopoly behavior evaluation index system; and the data target range is a monopoly behavior data target range.

9. The method according to claim 7, characterized in that: The prediction method also includes step four: obtaining the target economic entity information to be monitored and the target economic environment information of the industry in which it is located, preprocessing the target economic entity information and the target economic environment information, using the digital twin prediction model to perform dynamic monitoring, and outputting the monopoly situation assessment results, and sending the assessment results to the supervision platform through the communication module.

Images