Scene-based data asset assessment method and system

Abstract

The invention relates to a scene-based data asset evaluation method and system. The method comprises the following steps: step 1, evaluating a scene in which data assets can be used; 2, calculating the contribution degree of a single determined scene; 3, single evaluation is carried out to determine the number of scenes; 4, calculating a scene correction coefficient; 5, evaluating a data asset value; according to the method, evaluation is carried out by adopting the scene number, the scene contribution degree and the correction coefficient, the requirements and evaluation of a cost method, a benefit method and a market method for data quality are reduced, the core of data asset evaluation is powerfully grasped, and an accurate and reliable basis is provided for data asset evaluation.

Description

Technical Field

[0001] The present invention relates to a data asset evaluation method and system for a scenario, belongs to the field of asset evaluation, and is mainly used to provide accurate and reliable evaluation information for data assets. Background Art

[0002] Data assets are infinitely replicable, cost-free, and applicable in multiple scenarios. When evaluating data assets using the existing income approach, the applicability of the income approach is considered based on the historical and future application prospects of the data assets, combined with the operating conditions of the businesses that use or intend to use the data assets, and with a focus on analyzing the predictability of the economic returns of the data assets. The historical application of data assets does not represent future returns, and for data assets used in multiple transactions, it is impossible to trace their previous application.

[0003] When evaluating data assets using the existing cost method, the applicability of the cost method should be considered, based on the total input required to form the data assets, by analyzing the correlation between the value of the data assets and the costs; the replacement cost of the data assets should be determined, including upfront expenses, direct costs, indirect costs, opportunity costs, and related taxes and fees; this method directly ignores the characteristics of data assets and only calculates various upfront costs. It is powerless for data assets after transactions or when the cost of data assets in continued use is almost zero.

[0004] When evaluating data assets using existing market-based approaches, the applicability of market-based approaches should be considered, including whether there is a legal, compliant, and active public trading market for the data asset or similar data assets, as well as the availability of an appropriate number of comparable cases. Appropriate comparable cases should be selected based on the characteristics of the data asset. Market-based approaches require reference to similar cases, but due to the nascent stage of data asset trading, there are relatively few reference examples, making market-based approaches difficult to assess.

[0005] In view of the shortcomings of the above three methods, a data asset evaluation method that can be combined with scenario applications is proposed. Summary of the Invention

[0006] The present invention proposes a scenario-based data asset evaluation method and system, which solves the technical problem of a fusion method based on the contribution of data assets in the scenario, the number of scenarios used, and the scenario correction coefficient, and systematically solves a series of problems existing in the current cost method, income method, and market method in data asset evaluation.

[0007] In order to solve the above-mentioned technical problems, the present invention adopts the following solutions:

[0008] The scenario-based data asset assessment method includes the following steps:

[0009] Step 1: Evaluate the scenarios in which data assets can be used;

[0010] Step 2: Calculate the contribution of a single determined scenario;

[0011] Step 3: Evaluate the number of single scenarios;

[0012] Step 4: Calculate the scene correction coefficient;

[0013] Step 5: Evaluate the value of data assets.

[0014] Furthermore, the step 1 of evaluating the scenarios in which data assets can be used is performed by the following steps:

[0015] S11. Enter the data asset name and search in the data scenario database;

[0016] S12. The search results show the specific scenario name, the problem solved by the scenario, the number of scenarios, the industry name, and the industry name. The actual result is the number of known scenarios.

[0017] S13. Determine specific scenarios in which the assessed entity can implement its capabilities.

[0018] Furthermore, the method for establishing the data scenario database in step 1 is as follows:

[0019] S21. Data asset classification;

[0020] S22. Process the data scenarios by arranging, summarizing, and classifying them;

[0021] S23. Establish the corresponding relationship between data assets and data scenarios;

[0022] S24. Set up a dual search technology route based on data asset name and data scenario;

[0023] S25. Auxiliary indicator search sets up national economic industry search.

[0024] Furthermore, in step S21, data asset classification considers both the integrity and divisibility of data assets. If a data asset can be divided into multiple data assets corresponding to different data scenarios, it is considered divisible. If a data asset is divided into multiple data assets but still corresponds to the same data scenario, it is considered non-divisible. The classification and naming of data assets takes into account the principle of minimum unit data, focusing on implementing specific scenario solutions.

[0025] Furthermore, in the step S22, data scenario classification is mainly based on solving specific problems. If the problem contains multiple sub-problems, if the corresponding data assets can be specifically subdivided, the data scenario will be further subdivided. If not, the data scenario will not be analyzed further.

[0026] Furthermore, in step S23, if a certain data asset solves multiple scenarios, it corresponds to multiple specific scenarios; if multiple data assets jointly solve a specific data scenario problem, it is necessary to calculate the weights of the data assets in the specific scenarios separately.

[0027] Furthermore, in order to improve computing efficiency and shorten the time used to judge data assets and data scenarios when the program is running, step S23 adopts a one-to-one correspondence or one-to-many, many-to-one, many-to-many and other correspondences between data assets and data scenarios in combination with spatial dimension judgment. According to the research area, data assets and data scenario public data source factors, the computer program only searches within one of the areas within the province or city, thereby improving retrieval and correspondence efficiency.

[0028] Furthermore, in the step S25, the industry name and the industrial name are set so that the data assets can be operated in the industry or industrial scenario.

[0029] Furthermore, in step 2, the contribution of a data asset in a single, specific scenario is calculated. The contribution refers to the difference between the data asset before and after use in the data scenario. Contribution calculations are primarily used to reduce costs, improve efficiency, identify risks, and other aspects.

[0030] The contribution of different data assets to different dimensions (individual, enterprise, society, and environment) is calculated by subtracting the previous value from the final value.

[0031] Determining specific scenarios refers to scenarios that the data asset subject has the ability to implement. Scenarios that can be implemented in the future or have not been developed by the data asset subject are not included.

[0032] Taking into account the multi-scenario usage characteristics of data assets, the sum of the calculated contributions of multiple single scenarios is the initial assessed value of the data asset.

[0033] The initial evaluation formula for data assets is:

[0034] Where n represents the available scenarios, and y represents the contribution in a single scenario.

[0035] The numerical calculation formula for the contribution of a single scene is: Y = y later value - y earlier value.

[0036] The contribution of a single scenario in the second year is the difference between the value generated by the second use of the data asset and the value generated by the first use of the data asset, and so on.

[0037] The contribution of a single scenario in the third year is the difference between the final value generated by the third use of the data asset and the previous value generated by the second use of the data asset, and so on.

[0038] The contribution of a single scenario in the nth year is the difference between the value generated by the nth use of the data asset and the previous value generated by the n-1th use of the data asset, and so on.

[0039] Furthermore, the method for establishing the data scenario contribution database in step 2 is as follows:

[0040] S31. Establish different problem scenarios based on the data asset classification in S21;

[0041] S32. Collect and organize the values before use;

[0042] S33. Calculate the contribution value of the data asset to the scenario after use;

[0043] S34, calculating the value after use minus the value before use;

[0044] S35. Set enterprise, industry, and sector attribute indicators to match scenario contributions.

[0045] Furthermore, in step 3, the number of scenarios that can be used in the entire life cycle of a single determined scenario data asset is evaluated;

[0046] First, determine the life cycle of the data assets to be evaluated;

[0047] Secondly, evaluate the sum of the number of scenarios that the data asset owner can license for transactions and internal use within the future lifecycle of the data asset. Licensed transactions refer to the use of data assets after legally required platform transactions, recorded as Xa, while internal use refers to the use of data assets within the group's internal companies, recorded as Xb. For example, if Company A primarily uses data assets for marketing and management purposes, the number of usage scenarios is 1. If Company A is a group company and can use data assets in branches or subsidiaries in other provinces in addition to its own province, the number of scenarios is the number of branches or subsidiaries.

[0048] If Enterprise A, in addition to its own use, authorizes other enterprises to use the data through a data exchange or industry data trading platform, the number of authorized trading enterprises is the number of scenarios.

[0049] The number of scenarios in the above three situations can be added up to the total number of scenarios under this marketing management scenario.

[0050] Xn=Xa+Xb, where Xa represents the number of transactions in the corresponding scenario of the data asset, and Xb represents the number of internal uses of the corresponding scenario of the data asset.

[0051] Finally, referring to the number of scenarios that can be used in different life cycles, a scenario quantity database was established. The scenario quantity database records the actual number of application scenarios of data assets in different cycles in various enterprises.

[0052] Furthermore, the method for establishing the scene quantity database in step 3 is as follows:

[0053] S41. Set database indicators based on data asset naming, the entire life cycle of data assets, and the number of scenarios in different years for a single scenario;

[0054] S42. Input indicator data;

[0055] S43. Input the data asset name to automatically calculate the average number of scenarios in which the data can be used.

[0056] Furthermore, the method for establishing the scene correction coefficient database in step 4 is as follows:

[0057] S51. Data collation, recording annual inputs: funds, personnel, market competition, business model, enterprise scale and other indicators;

[0058] S52. Record the actual number of scenario-based uses in a single scenario per year;

[0059] S53. Use neural network model training to obtain a data scenario quantity model for the next year;

[0060] S54. Input data to calculate the scenario correction coefficient for the next year.

[0061] Further in S51, data such as capital, personnel, market competition, and enterprise scale are fused from multiple sources and input into a convolutional neural network to predict the usage quantity of the scenario in the next year.

[0062] The model training method is as follows:

[0063] This convolutional neural network consists of two convolution kernels: one for digital archive requirements and one for physical archives. Each layer is connected in series. The convolution layer consists of two 5x5 convolution kernels, or two-dimensional matrices, that perform convolution operations on the input features of the convolution layer. The maximum pooling layer calculates the maximum value of each neighborhood and passes it to the next layer. The regression layer outputs the target's location and category number. This neural network has a simple structure, low computational complexity, and high computational speed.

[0064] The trained convolutional neural network model is obtained by training the labeled historical data sample set. During the training process, the convolution kernel parameters in the convolutional neural network are optimized using the stochastic gradient descent method so that the value of the loss function converges to a preset value, making the loss function as small as possible and improving the accuracy of target detection.

[0065] The training process is as follows: the marked historical data x is used as the input of the convolutional neural network, and the target position y = f(x) under the current network parameters is obtained through layer-by-layer calculation of the neural network f. The target is the predicted value of the number of scenes yp = f(xp) and the actual number of scenes required value yc = f(xc).

[0066] The time of the historical data marked in the corresponding data x label is used as the actual target position The predicted values of the number of scenes are and actual scenario quantity requirements Calculate the predicted value y of the scenario quantity and the actual scenario quantity demand value respectively The error between .

[0067] This technical solution establishes the S51 prediction value and the S52 actual scenario quantity demand value; the data scenario demand data is implemented in the following way.

[0068] In this solution, the loss function loss is the difference between the scene data inference data and the actual scene quantity required value. Weighted sum of errors:

[0069]

[0070] Among them, α coord and α noobj are the target and actual weights respectively. In this solution, the target is specifically the impact scenario data; If it is a target, the item is 1, otherwise it is 0; If the target is present, this item is 0; otherwise, it is 1. x, y, w, and h are the target's funding, personnel, market competition, and enterprise size, respectively. C is the number of target scenarios, and p(c) is the probability that the target is of type c. Symbols with a ^ represent the network's estimated value for the corresponding attribute.

[0071] The gradient descent method is used to update the parameters θ of the neural network, that is,

[0072]

[0073] Where α is the learning rate, which is generally not greater than 0.01. In this embodiment, α=10 -5 .

[0074] The stochastic gradient descent method is used to optimize the convolution kernel parameters in the convolutional neural network so that the value of the loss function converges to a preset value. The value of the loss function converges to the preset value. The value of the loss function is required to be less than 0.6. After convergence, the value of the loss function no longer changes significantly with the increase of the number of iterations.

[0075] In this embodiment, the data scenario quantity loss function is calculated. When the values of the demand prediction value loss function and the actual demand value loss function converge to below the preset value of 0.3, the iterative training is terminated to generate the trained convolutional neural network model.

[0076] Furthermore, S54 inputs data to calculate the data asset scenario correction coefficient:

[0077] Input the scenario data into the S53 model to obtain the correction coefficient for the next year of the scenario;

[0078] Furthermore, an industry scenario correction coefficient database is established as follows:

[0079] When establishing the industry scenario correction coefficient, factors such as enterprise size, market share, and competitor industry development cycle are considered.

[0080] S61. Record: indicators such as enterprise size, market share, competitors, and industry development cycle;

[0081] S62. Record the actual usage in different scenarios in the industry;

[0082] S63. Use neural network model training to obtain the quantitative model of industry data scenarios for the next year;

[0083] S64. Input data to calculate the industry scenario correction coefficient for the next year.

[0084] Further in S61, data such as enterprise size, market share, competitors, and industry development cycle are fused from multiple sources and input into the convolutional neural network to predict the usage quantity of the scenario in the next year.

[0085] The model training method is as follows:

[0086] This convolutional neural network consists of two convolution kernels: one for digital archive requirements and one for physical archives. Each layer is connected in series. The convolution layer consists of two 5x5 convolution kernels, or two-dimensional matrices, that perform convolution operations on the input features of the convolution layer. The maximum pooling layer calculates the maximum value of each neighborhood and passes it to the next layer. The regression layer outputs the target's location and category number. This neural network has a simple structure, low computational complexity, and high computational speed.

[0087] The trained convolutional neural network model is obtained by training the labeled historical data sample set. During the training process, the convolution kernel parameters in the convolutional neural network are optimized using the stochastic gradient descent method so that the value of the loss function converges to a preset value, making the loss function as small as possible and improving the accuracy of target detection.

[0088] The training process is as follows: the marked historical data x is used as the input of the convolutional neural network, and the target position y = f(x) under the current network parameters is obtained through layer-by-layer calculation of the neural network f. The target is the predicted value of the number of scenes yp = f(xp) and the actual number of scenes required value yc = f(xc).

[0089] The time of the historical data marked in the corresponding data x label is used as the actual target position The predicted values of the number of scenes are and actual scenario quantity requirements Calculate the predicted value y of the scenario quantity and the actual scenario quantity demand value respectively The error between .

[0090] This technical solution establishes the S61 prediction value and the S62 actual scenario quantity requirement value; the data scenario requirement data is implemented in the following way.

[0091] In this solution, the loss function loss is the difference between the scene data inference data and the actual scene quantity required value. Weighted sum of errors:

[0092]

[0093] Among them, α coord and α noobj are the target and actual weights respectively. In this solution, the target is specifically the impact scenario data; If it is a target, the item is 1, otherwise it is 0; If the target is present, this item is 0; otherwise, it is 1. x, y, w, and h are the target's enterprise size, market share, competitors, and industry development cycle, respectively. C is the number of target scenarios, and p(c) is the probability that the target is of type c. Symbols with a ^ represent the network's estimated value for the corresponding attribute.

[0094] The gradient descent method is used to update the parameters θ of the neural network, that is,

[0095]

[0096] Where α is the learning rate, which is generally not greater than 0.01. In this embodiment, α=10 -5 .

[0097] The stochastic gradient descent method is used to optimize the convolution kernel parameters in the convolutional neural network so that the value of the loss function converges to a preset value. The value of the loss function converges to the preset value. The value of the loss function is required to be less than 0.6. After convergence, the value of the loss function no longer changes significantly with the increase of the number of iterations.

[0098] In this embodiment, the data scenario quantity loss function is calculated. When the values of the demand prediction value loss function and the actual demand value loss function converge to below the preset value of 0.3, the iterative training is terminated to generate the trained convolutional neural network model.

[0099] Furthermore, S64 inputs data to calculate the data asset scenario correction coefficient:

[0100] Input the scenario data into the S63 model to obtain the correction coefficient for the next year of the scenario.

[0101] Furthermore, the step 5 evaluates the value of data assets.

[0102] P0=Y0

[0103] P=Y1×X1×K1+Y2×X2×K2+…+Yn×Xn×Kn

[0104] Among them, Yn represents the contribution value of the nth single scene, Xn represents the number of the nth single scene, and Kn represents the correction coefficient of the nth single scene.

[0105] The contribution of a single scenario in the second year is the difference between the value generated by the second use of the data asset and the value generated by the first use of the data asset, and so on.

[0106] The contribution of a single scenario in the third year is the difference between the final value generated by the third use of the data asset and the previous value generated by the second use of the data asset, and so on.

[0107] The contribution of a single scenario in the nth year is the difference between the value generated by the nth use of the data asset and the previous value generated by the n-1th use of the data asset, and so on.

[0108] Formula for initial valuation of data assets:

[0109] Where n represents the available scenarios, and y represents the contribution in a single scenario; it can be used to evaluate the initial value of data assets, that is, the valuation of data assets in the first year.

[0110] The initial value of the scenario correction coefficient K1 corresponding to Y1 is 1. The ratio of the actual number of usage scenarios in the two years after the evaluation to the number of usage scenarios during the evaluation is K12, which serves as the scenario correction coefficient for the second year under the user entity; it is used in the data asset evaluation formula for the second year.

[0111] The ratio of the actual number of usage scenarios in the three years after the evaluation to the number of usage scenarios during the evaluation is K13, which serves as the scenario correction coefficient for the third year under the user entity; it is used in the data asset evaluation formula for the third year.

[0112] The ratio of the actual number of usage scenarios in the four years after the evaluation to the number of usage scenarios during the evaluation is K14, which serves as the scenario correction coefficient for the fourth year under the user entity; it is used in the fourth-year data asset evaluation formula.

[0113] The initial value of the scenario correction coefficient K2 corresponding to Y2 is 1. The ratio of the actual number of usage scenarios in the two years after the evaluation to the number of usage scenarios during the evaluation is K22, which serves as the scenario correction coefficient for the second year under the user entity; it is used in the data asset evaluation formula for the second year.

[0114] The ratio of the actual number of usage scenarios in the three years after the evaluation to the number of usage scenarios during the evaluation is K23, which serves as the scenario correction coefficient for the third year under the user entity; it is used in the data asset evaluation formula for the third year.

[0115] The ratio of the actual number of usage scenarios in the four years after the evaluation to the number of usage scenarios during the evaluation is K24, which serves as the scenario correction coefficient for the fourth year under the user entity; it is used in the fourth-year data asset evaluation formula.

[0116] The initial value of the scenario correction coefficient Kn corresponding to Yn is 1. The ratio of the actual number of usage scenarios in the two years after the evaluation to the number of usage scenarios during the evaluation is Kn2, which serves as the scenario correction coefficient for the second year under the user entity; it is used in the data asset evaluation formula for the second year.

[0117] The ratio of the actual number of usage scenarios in the three years after the evaluation to the number of usage scenarios during the evaluation is Kn3, which serves as the scenario correction coefficient for the third year under the user entity; it is used in the data asset evaluation formula for the third year.

[0118] The ratio of the actual number of usage scenarios in the four years after the evaluation to the number of usage scenarios during the evaluation is Kn4, which serves as the scenario correction coefficient for the fourth year under the user entity; it is used in the fourth-year data asset evaluation formula.

[0119] The second year evaluation formula is:

[0120] P=Y1×X1×K12+Y2×X2×K22+…+Yn×Xn×Kn2

[0121] The third year evaluation formula is:

[0122] P=Y1×X1×K13+Y2×X2×K23+…+Yn×Xn×Kn3

[0123] The fourth year evaluation formula is:

[0124] P=Y1×X1×K14+Y2×X2×K24+…+Yn×Xn×Kn4

[0125] Evaluation formula for the Nth year:

[0126] P=Y1×X1×K1n+Y2×X2×K2n+…+Yn×Xn×Knn

[0127] If the contribution is decreasing, you can recalculate the contribution and substitute it into the above formula:

[0128] The contribution of a single scenario in the second year is the difference between the value generated by the second use of the data asset and the value generated by the first use of the data asset, and so on.

[0129] The contribution of a single scenario in the third year is the difference between the final value generated by the third use of the data asset and the previous value generated by the second use of the data asset, and so on.

[0130] The contribution of a single scenario in the nth year is the difference between the value generated by the nth use of the data asset and the previous value generated by the n-1th use of the data asset, and so on.

[0131] The present invention also includes a scenario-based data asset evaluation system, which includes a data scenario database module, a contribution calculation module, a scenario quantity database module, a scenario correction coefficient database module, an evaluation module, and an evaluation report generation module.

[0132] Data scenario database module: used to query the scenarios in which data assets are used, establish the correspondence between data asset reclassification and data scenarios or single specific problems, grasp the range of known scenarios in which data assets may be used as much as possible, and establish scenario innovation methods for unknown scenarios to generate new data scenarios for the data assets.

[0133] Contribution calculation module: used to calculate the contribution of a single scenario in solving a certain problem. Scenario-based solutions mainly focus on reducing costs, improving efficiency, identifying risks, and other aspects. The above three aspects are quantified. For cost reduction, the absolute value of the difference between the cost when data assets are not used and the cost when data assets are used is the contribution. When calculating the contribution of cost reduction, the contribution of the simple reduction of data assets should be considered. Since there are multiple factors that lead to cost reduction, the module sets different weights for each factor of cost reduction, which can be adjusted by the user according to actual conditions to calculate the contribution value brought by the single factor, that is, the use of data assets.

[0134] Efficiency improvement is achieved by multiplying the decision-making time reduction by the decision-making effectiveness coefficient. The contribution is calculated by combining the decision-making team's overall daily salary and the time allocated to the decision-making process. For example, if the use of data assets reduces decision-making efficiency or approval process time by two days, the decision-making team's salary for those two days will be used as the contribution value. If the decision-making accuracy is greater than 50-100%, the decision-making effectiveness coefficient is set between 2 and 10, and the user will determine this based on their experience.

[0135] The added value of benefits from identifying and eliminating risks. The benefits of identifying and eliminating risks driven by a single data asset are calculated as the short-term losses incurred while the risk exists plus the short-term benefits from normal production operations.

[0136] Other aspects can be set by users. If they cannot be attributed to the above three aspects, the user can set the income contribution value brought by the data assets.

[0137] Scenario quantity database module: used for calculating the number of replicable usage scenarios of a single scenario in different life cycle years for multiple scenarios within the subject control cycle of data assets.

[0138] Scenario correction coefficient database module: used for scenario correction coefficients in different industries; factors such as enterprise size, personnel, financial strength, market share, number of enterprises, and industry development cycle are considered when establishing industry scenario correction coefficients; at the same time, the influencing factors of the enterprise's internal use of data scenarios are also used to calculate enterprise correction coefficients in different years, as well as industry scenario correction coefficients.

[0139] Evaluation module: used to evaluate the value of the data asset, based on the data scenario evaluation formula:

[0140] P0=Y0

[0141] P=Y1×X1×K1+Y2×X2×K2+…+Yn×Xn×Kn

[0142] Calculate the value of the data asset.

[0143] Where Yn represents the contribution value of the nth single scenario, Xn represents the number of nth single scenarios, and Kn represents the correction coefficient of the nth single scenario. The contribution value of a single scenario in the second year is the difference between the value generated by the second use of the data asset and the value generated by the first use of the data asset, and so on.

[0144] The contribution of a single scenario in the third year is the difference between the final value generated by the third use of the data asset and the previous value generated by the second use of the data asset, and so on.

[0145] The contribution of a single scenario in the nth year is the difference between the value generated by the nth use of the data asset and the previous value generated by the n-1th use of the data asset, and so on.

[0146] Formula for initial valuation of data assets:

[0147] Where n represents the available scenarios, and y represents the contribution in a single scenario; it can be used to evaluate the initial value of data assets, that is, the valuation of data assets in the first year.

[0148] The initial value of the scenario correction coefficient K1 corresponding to Y1 is 1. The ratio of the actual number of usage scenarios in the two years after the evaluation to the number of usage scenarios during the evaluation is K12, which serves as the scenario correction coefficient for the second year under the user entity; it is used in the data asset evaluation formula for the second year.

[0149] The ratio of the actual number of usage scenarios in the three years after the evaluation to the number of usage scenarios during the evaluation is K13, which serves as the scenario correction coefficient for the third year under the user entity; it is used in the data asset evaluation formula for the third year.

[0150] The ratio of the actual number of usage scenarios in the four years after the evaluation to the number of usage scenarios during the evaluation is K14, which serves as the scenario correction coefficient for the fourth year under the user entity; it is used in the fourth-year data asset evaluation formula.

[0151] The initial value of the scenario correction coefficient K2 corresponding to Y2 is 1. The ratio of the actual number of usage scenarios in the two years after the evaluation to the number of usage scenarios during the evaluation is K22, which serves as the scenario correction coefficient for the second year under the user entity; it is used in the data asset evaluation formula for the second year.

[0152] The ratio of the actual number of usage scenarios in the three years after the evaluation to the number of usage scenarios during the evaluation is K23, which serves as the scenario correction coefficient for the third year under the user entity; it is used in the data asset evaluation formula for the third year.

[0153] The ratio of the actual number of usage scenarios in the four years after the evaluation to the number of usage scenarios during the evaluation is K24, which serves as the scenario correction coefficient for the fourth year under the user entity; it is used in the fourth-year data asset evaluation formula.

[0154] The initial value of the scenario correction coefficient Kn corresponding to Yn is 1. The ratio of the actual number of usage scenarios in the two years after the evaluation to the number of usage scenarios during the evaluation is Kn2, which serves as the scenario correction coefficient for the second year under the user entity; it is used in the data asset evaluation formula for the second year.

[0155] The ratio of the actual number of usage scenarios in the three years after the evaluation to the number of usage scenarios during the evaluation is Kn3, which serves as the scenario correction coefficient for the third year under the user entity; it is used in the data asset evaluation formula for the third year.

[0156] The ratio of the actual number of usage scenarios in the four years after the evaluation to the number of usage scenarios during the evaluation is Kn4, which serves as the scenario correction coefficient for the fourth year under the user entity; it is used in the fourth-year data asset evaluation formula.

[0157] The second year evaluation formula is:

[0158] P=Y1×X1×K12+Y2×X2×K22+…+Yn×Xn×Kn2

[0159] The third year evaluation formula is:

[0160] P=Y1×X1×K13+Y2×X2×K23+…+Yn×Xn×Kn3

[0161] The fourth year evaluation formula is:

[0162] P=Y1×X1×K14+Y2×X2×K24+…+Yn×Xn×Kn4

[0163] Evaluation formula for the Nth year:

[0164] P=Y1×X1×K1n+Y2×X2×K2n+…+Yn×Xn×Knn

[0165] If the contribution is decreasing, you can recalculate the contribution and substitute it into the above formula:

[0166] The contribution of a single scenario in the second year is the difference between the value generated by the second use of the data asset and the value generated by the first use of the data asset, and so on.

[0167] The contribution of a single scenario in the third year is the difference between the final value generated by the third use of the data asset and the previous value generated by the second use of the data asset, and so on.

[0168] The contribution of a single scenario in the nth year is the difference between the value generated by the nth use of the data asset and the previous value generated by the n-1th use of the data asset, and so on.

[0169] Evaluation report generation module: used to generate evaluation reports and list the data scenarios used by the data asset, unused data scenarios, evaluation valuation, scenario correction coefficient, and number of scenarios.

[0170] The data asset evaluation method and system of the present invention have the following beneficial effects:

[0171] (1) This invention uses the number of scenarios and the degree of scenario contribution for evaluation, which reduces the requirements and evaluations of data quality by the cost method, benefit method, and market method, and effectively grasps the core of data asset evaluation.

[0172] (2) The present invention adopts the sum of the contributions of multiple scenarios, taking into account the practical problems of multi-scenario applications of a single data asset, and gives higher value to the data asset.

[0173] (3) The scenario coefficient correction in the present invention makes it possible to use multiple scenarios and to supplement the actual use of multiple scenarios, reducing the number of unrealizable scenarios in the evaluation, and is more in line with the requirement that the value of data asset evaluation depends on the ability of the user, providing an accurate and reliable basis for data asset evaluation.

[0174] (4) The present invention uses a neural network to perform scenario-based predictions on factors that actually affect usage scenarios, and trains the model's accuracy in the use of scenario correction coefficients.

[0175] (5) The present invention adopts a data asset and data scenario database, a scenario quantity database, and a scenario correction coefficient database to improve the accuracy of data asset evaluation.

[0176] (6) The present invention adopts the enterprise data scenario correction coefficient to form the industry scenario correction coefficient, which is conducive to the input of the initial value of the data scenario coefficient according to the industry in the future. BRIEF DESCRIPTION OF THE DRAWINGS

[0177] Figure 1 : The present invention provides a scenario-based data asset evaluation method and system flow chart. DETAILED DESCRIPTION

[0178] The present invention proposes a scenario-based data asset evaluation method and system, which solves the technical problem of a fusion method based on the contribution of data assets in the scenario, the number of scenarios used, and the scenario correction coefficient, and systematically solves a series of problems existing in the current cost method, income method, and market method in data asset evaluation.

[0179] In order to solve the above-mentioned technical problems, the present invention adopts the following solutions:

[0180] The scenario-based data asset assessment method includes the following steps:

[0181] Step 1: Evaluate the scenarios in which data assets can be used;

[0182] Step 2: Calculate the contribution of a single determined scenario;

[0183] Step 3: Evaluate the number of single scenarios;

[0184] Step 4: Calculate the scene correction coefficient;

[0185] Step 5: Evaluate the value of data assets.

[0186] Furthermore, the step 1 of evaluating the scenarios in which data assets can be used is performed by the following steps:

[0187] S11. Enter the data asset name and search in the data scenario database;

[0188] S12. The search results show the specific scenario name, the problem solved by the scenario, the number of scenarios, the industry name, and the industry name. The actual result is the number of known scenarios.

[0189] S13. Determine specific scenarios in which the assessed entity can implement its capabilities.

[0190] Furthermore, the method for establishing the data scenario database in step 1 is as follows:

[0191] S21. Data asset classification;

[0192] S22. Process the data scenarios by arranging, summarizing, and classifying them;

[0193] S23. Establish the corresponding relationship between data assets and data scenarios;

[0194] S24. Set up a dual search technology route based on data asset name and data scenario;

[0195] S25. Auxiliary indicator search sets up national economic industry search.

[0196] Furthermore, in step S21, data asset classification considers both the integrity and divisibility of data assets. If a data asset can be divided into multiple data assets corresponding to different data scenarios, it is considered divisible. If a data asset is divided into multiple data assets but still corresponds to the same data scenario, it is considered non-divisible. The classification and naming of data assets takes into account the principle of minimum unit data, focusing on implementing specific scenario solutions.

[0197] Furthermore, in the step S22, data scenario classification is mainly based on solving specific problems. If the problem contains multiple sub-problems, if the corresponding data assets can be specifically subdivided, the data scenario will be further subdivided. If not, the data scenario will not be analyzed further.

[0198] Furthermore, in step S23, if a certain data asset solves multiple scenarios, it corresponds to multiple specific scenarios; if multiple data assets jointly solve a specific data scenario problem, it is necessary to calculate the weights of the data assets in the specific scenarios separately.

[0199] Furthermore, in order to improve computing efficiency and shorten the time used to judge data assets and data scenarios when the program is running, step S23 adopts a one-to-one correspondence or one-to-many, many-to-one, many-to-many and other correspondences between data assets and data scenarios in combination with spatial dimension judgment. According to the research area, data assets and data scenario public data source factors, the computer program only searches within one of the areas within the province or city, thereby improving retrieval and correspondence efficiency.

[0200] Furthermore, in the step S25, the industry name and the industrial name are set so that the data assets can be operated in the industry or industrial scenario.

[0201] Furthermore, in step 2, the contribution of a data asset in a single, specific scenario is calculated. The contribution refers to the difference between the data asset before and after use in the data scenario. Contribution calculations are primarily used to reduce costs, improve efficiency, identify risks, and other aspects.

[0202] The contribution of different data assets to different dimensions (individual, enterprise, society, and environment) is calculated by subtracting the previous value from the final value.

[0203] Determining specific scenarios refers to scenarios that the data asset subject has the ability to implement. Scenarios that can be implemented in the future or have not been developed by the data asset subject are not included.

[0204] Taking into account the multi-scenario usage characteristics of data assets, the sum of the calculated contributions of multiple single scenarios is the initial assessed value of the data asset.

[0205] The initial evaluation formula for data assets is:

[0206] Where n represents the available scenarios, and y represents the contribution in a single scenario.

[0207] The numerical calculation formula for the contribution of a single scene is: Y = y later value - y earlier value.

[0208] The contribution of a single scenario in the second year is the difference between the value generated by the second use of the data asset and the value generated by the first use of the data asset, and so on.

[0209] The contribution of a single scenario in the third year is the difference between the final value generated by the third use of the data asset and the previous value generated by the second use of the data asset, and so on.

[0210] The contribution of a single scenario in the nth year is the difference between the value generated by the nth use of the data asset and the previous value generated by the n-1th use of the data asset, and so on.

[0211] Efficiency improvement is achieved by multiplying the decision-making time reduction by the decision-making effectiveness coefficient. The contribution is calculated by combining the decision-making team's overall daily salary and the time allocated to the decision-making process. For example, if the use of data assets reduces decision-making efficiency or approval process time by two days, the decision-making team's salary for those two days will be used as the contribution value. If the decision-making accuracy is greater than 50-100%, the decision-making effectiveness coefficient is set between 2 and 10, and the user will determine this based on their experience.

[0212] The added value of benefits from identifying and eliminating risks. The benefits of identifying and eliminating risks driven by a single data asset are calculated as the short-term losses incurred while the risk exists plus the short-term benefits from normal production operations.

[0213] Other aspects can be set by users. If they cannot be attributed to the above three aspects, the user can set the income contribution value brought by the data assets.

[0214] Furthermore, the method for establishing the data scenario contribution database in step 2 is as follows:

[0215] S31. Establish different problem scenarios based on the data asset classification in S21;

[0216] S32. Collect and organize the values before use;

[0217] S33. Calculate the contribution value of the data asset to the scenario after use;

[0218] S34, calculating the value after use minus the value before use;

[0219] S35. Set enterprise, industry, and sector attribute indicators to match scenario contributions.

[0220] Furthermore, in step 3, the number of scenarios that can be used in the entire life cycle of a single determined scenario data asset is evaluated;

[0221] First, determine the life cycle of the data assets to be evaluated;

[0222] Secondly, evaluate the sum of the number of scenarios that the data asset owner can license for transactions and internal use within the future lifecycle of the data asset. Licensed transactions refer to the use of data assets after legally required platform transactions, recorded as Xa, while internal use refers to the use of data assets within the group's internal companies, recorded as Xb. For example, if Company A primarily uses data assets for marketing and management purposes, the number of usage scenarios is 1. If Company A is a group company and can use data assets in branches or subsidiaries in other provinces in addition to its own province, the number of scenarios is the number of branches or subsidiaries.

[0223] If Enterprise A, in addition to its own use, authorizes other enterprises to use the data through a data exchange or industry data trading platform, the number of authorized trading enterprises is the number of scenarios.

[0224] The number of scenarios in the above three situations can be added up to the total number of scenarios under this marketing management scenario.

[0225] Xn=Xa+Xb, where Xa represents the number of transactions in the corresponding scenario of the data asset, and Xb represents the number of internal uses of the corresponding scenario of the data asset.

[0226] Finally, referring to the number of scenarios that can be used in different life cycles, a scenario quantity database was established. The scenario quantity database records the actual number of application scenarios of data assets in different cycles in various enterprises.

[0227] Furthermore, the method for establishing the scene quantity database in step 3 is as follows:

[0228] S41. Set database indicators based on data asset naming, the entire life cycle of data assets, and the number of scenarios in different years for a single scenario;

[0229] S42. Input indicator data;

[0230] S43. Input the data asset name to automatically calculate the average number of scenarios in which the data can be used.

[0231] Furthermore, the method for establishing the scene correction coefficient database in step 4 is as follows:

[0232] S51. Data collation, recording annual inputs: funds, personnel, market competition, business model, enterprise scale and other indicators;

[0233] S52. Record the actual number of scenario-based uses in a single scenario per year;

[0234] S53. Use neural network model training to obtain a data scenario quantity model for the next year;

[0235] S54. Input data to calculate the scenario correction coefficient for the next year.

[0236] Further in S51, data such as capital, personnel, market competition, and enterprise scale are fused from multiple sources and input into a convolutional neural network to predict the usage quantity of the scenario in the next year.

[0237] The model training method is as follows:

[0238] This convolutional neural network consists of two convolution kernels: one for digital archive requirements and one for physical archives. Each layer is connected in series. The convolution layer consists of two 5x5 convolution kernels, or two-dimensional matrices, that perform convolution operations on the input features of the convolution layer. The maximum pooling layer calculates the maximum value of each neighborhood and passes it to the next layer. The regression layer outputs the target's location and category number. This neural network has a simple structure, low computational complexity, and high computational speed.

[0239] The trained convolutional neural network model is obtained by training the labeled historical data sample set. During the training process, the convolution kernel parameters in the convolutional neural network are optimized using the stochastic gradient descent method so that the value of the loss function converges to a preset value, making the loss function as small as possible and improving the accuracy of target detection.

[0240] The training process is as follows: the marked historical data x is used as the input of the convolutional neural network, and the target position y = f(x) under the current network parameters is obtained through layer-by-layer calculation of the neural network f. The target is the predicted value of the number of scenes yp = f(xp) and the actual number of scenes required value yc = f(xc).

[0241] The time of the historical data marked in the corresponding data x label is used as the actual target position The predicted values of the number of scenes are and actual scenario quantity requirements Calculate the predicted value y of the scenario quantity and the actual scenario quantity demand value respectively The error between .

[0242] This technical solution establishes the S51 prediction value and the S52 actual scenario quantity demand value; the data scenario demand data is implemented in the following way.

[0243] In this solution, the loss function loss is the difference between the scene data inference data and the actual scene quantity required value. Weighted sum of errors:

[0244]

[0245] Among them, α coord and α noobj are the target and actual weights respectively. In this solution, the target is specifically the impact scenario data; If it is a target, the item is 1, otherwise it is 0; If the target is present, this item is 0; otherwise, it is 1. x, y, w, and h are the target's funding, personnel, market competition, and enterprise size, respectively. C is the number of target scenarios, and p(c) is the probability that the target is of type c. Symbols with a ^ represent the network's estimated value for the corresponding attribute.

[0246] The gradient descent method is used to update the parameters θ of the neural network, that is,

[0247]

[0248] Where α is the learning rate, which is generally not greater than 0.01. In this embodiment, α=10 -5 .

[0249] The stochastic gradient descent method is used to optimize the convolution kernel parameters in the convolutional neural network so that the value of the loss function converges to a preset value. The value of the loss function converges to the preset value. The value of the loss function is required to be less than 0.6. After convergence, the value of the loss function no longer changes significantly with the increase of the number of iterations.

[0250] In this embodiment, the data scenario quantity loss function is calculated. When the values of the demand prediction value loss function and the actual demand value loss function converge to below the preset value of 0.3, the iterative training is terminated to generate the trained convolutional neural network model.

[0251] Furthermore, S54 inputs data to calculate the data asset scenario correction coefficient:

[0252] Input the scenario data into the S53 model to obtain the correction coefficient for the next year of the scenario;

[0253] Furthermore, an industry scenario correction coefficient database is established as follows:

[0254] When establishing the industry scenario correction coefficient, factors such as enterprise size, market share, and competitor industry development cycle are considered.

[0255] S61. Record: indicators such as enterprise size, market share, competitors, and industry development cycle;

[0256] S62. Record the actual usage in different scenarios in the industry;

[0257] S63. Use neural network model training to obtain the quantitative model of industry data scenarios for the next year;

[0258] S64. Input data to calculate the industry scenario correction coefficient for the next year.

[0259] Further in S61, data such as enterprise size, market share, competitors, and industry development cycle are fused from multiple sources and input into the convolutional neural network to predict the usage quantity of the scenario in the next year.

[0260] The model training method is as follows:

[0261] This convolutional neural network consists of two convolution kernels: one for digital archive requirements and one for physical archives. Each layer is connected in series. The convolution layer consists of two 5x5 convolution kernels, or two-dimensional matrices, that perform convolution operations on the input features of the convolution layer. The maximum pooling layer calculates the maximum value of each neighborhood and passes it to the next layer. The regression layer outputs the target's location and category number. This neural network has a simple structure, low computational complexity, and high computational speed.

[0262] The trained convolutional neural network model is obtained by training the labeled historical data sample set. During the training process, the convolution kernel parameters in the convolutional neural network are optimized using the stochastic gradient descent method so that the value of the loss function converges to a preset value, making the loss function as small as possible and improving the accuracy of target detection.

[0263] The training process is as follows: the marked historical data x is used as the input of the convolutional neural network, and the target position y = f(x) under the current network parameters is obtained through layer-by-layer calculation of the neural network f. The target is the predicted value of the number of scenes yp = f(xp) and the actual number of scenes required value yc = f(xc).

[0264] The time of the historical data marked in the corresponding data x label is used as the actual target position The predicted values of the number of scenes are and actual scenario quantity requirements Calculate the predicted value y of the scenario quantity and the actual scenario quantity demand value respectively The error between .

[0265] This technical solution establishes the S61 prediction value and the S62 actual scenario quantity requirement value; the data scenario requirement data is implemented in the following way.

[0266] In this solution, the loss function loss is the difference between the scene data inference data and the actual scene quantity required value. Weighted sum of errors:

[0267]

[0268] Among them, α coord and α noobj are the target and actual weights respectively. In this solution, the target is specifically the impact scenario data; If it is a target, the item is 1, otherwise it is 0; If the target is present, this item is 0; otherwise, it is 1. x, y, w, and h are the target's enterprise size, market share, competitors, and industry development cycle, respectively. C is the number of target scenarios, and p(c) is the probability that the target is of type c. Symbols with a ^ represent the network's estimated value for the corresponding attribute.

[0269] The gradient descent method is used to update the parameters θ of the neural network, that is,

[0270]

[0271] Where α is the learning rate, which is generally not greater than 0.01. In this embodiment, α=10 -5 .

[0272] The stochastic gradient descent method is used to optimize the convolution kernel parameters in the convolutional neural network so that the value of the loss function converges to a preset value. The value of the loss function converges to the preset value. The value of the loss function is required to be less than 0.6. After convergence, the value of the loss function no longer changes significantly with the increase of the number of iterations.

[0273] In this embodiment, the data scenario quantity loss function is calculated. When the values of the demand prediction value loss function and the actual demand value loss function converge to below the preset value of 0.3, the iterative training is terminated to generate the trained convolutional neural network model.

[0274] Furthermore, S64 inputs data to calculate the data asset scenario correction coefficient:

[0275] Input the scenario data into the S63 model to obtain the correction coefficient for the next year of the scenario.

[0276] Furthermore, the step 5 evaluates the value of data assets.

[0277] P0=Y0

[0278] P=Y1×X1×K1+Y2×X2×K2+…+Yn×Xn×Kn

[0279] Among them, Yn represents the contribution value of the nth single scene, Xn represents the number of the nth single scene, and Kn represents the correction coefficient of the nth single scene;

[0280] The contribution of a single scenario in the second year is the difference between the value generated by the second use of the data asset and the value generated by the first use of the data asset, and so on.

[0281] The contribution of a single scenario in the third year is the difference between the final value generated by the third use of the data asset and the previous value generated by the second use of the data asset, and so on.

[0282] The contribution of a single scenario in the nth year is the difference between the value generated by the nth use of the data asset and the previous value generated by the n-1th use of the data asset, and so on.

[0283] Formula for initial valuation of data assets:

[0284] Where n represents the available scenarios, and y represents the contribution in a single scenario; it can be used to evaluate the initial value of data assets, that is, the valuation of data assets in the first year.

[0285] The initial value of the scenario correction coefficient K1 corresponding to Y1 is 1. The ratio of the actual number of usage scenarios in the two years after the evaluation to the number of usage scenarios during the evaluation is K12, which serves as the scenario correction coefficient for the second year under the user entity; it is used in the data asset evaluation formula for the second year.

[0286] The ratio of the actual number of usage scenarios in the three years after the evaluation to the number of usage scenarios during the evaluation is K13, which serves as the scenario correction coefficient for the third year under the user entity; it is used in the data asset evaluation formula for the third year.

[0287] The ratio of the actual number of usage scenarios in the four years after the evaluation to the number of usage scenarios during the evaluation is K14, which serves as the scenario correction coefficient for the fourth year under the user entity; it is used in the fourth-year data asset evaluation formula.

[0288] The initial value of the scenario correction coefficient K2 corresponding to Y2 is 1. The ratio of the actual number of usage scenarios in the two years after the evaluation to the number of usage scenarios during the evaluation is K22, which serves as the scenario correction coefficient for the second year under the user entity; it is used in the data asset evaluation formula for the second year.

[0289] The ratio of the actual number of usage scenarios in the three years after the evaluation to the number of usage scenarios during the evaluation is K23, which serves as the scenario correction coefficient for the third year under the user entity; it is used in the data asset evaluation formula for the third year.

[0290] The ratio of the actual number of usage scenarios in the four years after the evaluation to the number of usage scenarios during the evaluation is K24, which serves as the scenario correction coefficient for the fourth year under the user entity; it is used in the fourth-year data asset evaluation formula.

[0291] The initial value of the scenario correction coefficient Kn corresponding to Yn is 1. The ratio of the actual number of usage scenarios in the two years after the evaluation to the number of usage scenarios during the evaluation is Kn2, which serves as the scenario correction coefficient for the second year under the user entity; it is used in the data asset evaluation formula for the second year.

[0292] The ratio of the actual number of usage scenarios in the three years after the evaluation to the number of usage scenarios during the evaluation is Kn3, which serves as the scenario correction coefficient for the third year under the user entity; it is used in the data asset evaluation formula for the third year.

[0293] The ratio of the actual number of usage scenarios in the four years after the evaluation to the number of usage scenarios during the evaluation is Kn4, which serves as the scenario correction coefficient for the fourth year under the user entity; it is used in the fourth-year data asset evaluation formula.

[0294] The second year evaluation formula is:

[0295] P=Y1×X1×K12+Y2×X2×K22+…+Yn×Xn×Kn2

[0296] The third year evaluation formula is:

[0297] P=Y1×X1×K13+Y2×X2×K23+…+Yn×Xn×Kn3

[0298] The fourth year evaluation formula is:

[0299] P=Y1×X1×K14+Y2×X2×K24+…+Yn×Xn×Kn4

[0300] Evaluation formula for the Nth year:

[0301] P=Y1×X1×K1n+Y2×X2×K2n+…+Yn×Xn×Knn

[0302] If the contribution is decreasing, you can recalculate the contribution and substitute it into the above formula:

[0303] The contribution of a single scenario in the second year is the difference between the value generated by the second use of the data asset and the value generated by the first use of the data asset, and so on.

[0304] The contribution of a single scenario in the third year is the difference between the final value generated by the third use of the data asset and the previous value generated by the second use of the data asset, and so on.

[0305] The contribution of a single scenario in the nth year is the difference between the value generated by the nth use of the data asset and the previous value generated by the n-1th use of the data asset, and so on.

[0306] The present invention also includes a scenario-based data asset evaluation system, which includes a data scenario database module, a contribution calculation module, a scenario quantity database module, a scenario correction coefficient database module, an evaluation module, and an evaluation report generation module.

[0307] Data scenario database module: used to query the scenarios in which data assets are used, establish the correspondence between data asset reclassification and data scenarios or single specific problems, grasp the range of known scenarios in which data assets may be used as much as possible, and establish scenario innovation methods for unknown scenarios to generate new data scenarios for the data assets.

[0308] Contribution calculation module: used to calculate the contribution of a single scenario in solving a certain problem. Scenario-based solutions mainly focus on reducing costs, improving efficiency, identifying risks, and other aspects. The above three aspects are quantified. For cost reduction, the absolute value of the difference between the cost when data assets are not used and the cost when data assets are used is the contribution. When calculating the contribution of cost reduction, the contribution of the simple reduction of data assets should be considered. Since there are multiple factors that lead to cost reduction, the module sets different weights for each factor of cost reduction, which can be adjusted by the user according to actual conditions to calculate the contribution value brought by the single factor, that is, the use of data assets.

[0309] Efficiency improvement is achieved by multiplying the decision-making time reduction by the decision-making effectiveness coefficient. The contribution is calculated by combining the decision-making team's overall daily salary and the time allocated to the decision-making process. For example, if the use of data assets reduces decision-making efficiency or approval process time by two days, the decision-making team's salary for those two days will be used as the contribution value. If the decision-making accuracy is greater than 50-100%, the decision-making effectiveness coefficient is set between 2 and 10, and the user will determine this based on their experience.

[0310] The added value of benefits from identifying and eliminating risks. The benefits of identifying and eliminating risks driven by a single data asset are calculated as the short-term losses incurred while the risk exists plus the short-term benefits from normal production operations.

[0311] Other aspects can be set by users. If they cannot be attributed to the above three aspects, the user can set the income contribution value brought by the data assets.

[0312] Scenario quantity database module: used for calculating the number of replicable usage scenarios of a single scenario in different life cycle years for multiple scenarios within the subject control cycle of data assets.

[0313] Scenario correction coefficient database module: used for scenario correction coefficients in different industries; factors such as enterprise size, personnel, financial strength, market share, number of enterprises, and industry development cycle are considered when establishing industry scenario correction coefficients; at the same time, the influencing factors of the enterprise's internal use of data scenarios are also used to calculate enterprise correction coefficients in different years, as well as industry scenario correction coefficients.

[0314] Evaluation module: used to evaluate the value of the data asset, based on the data scenario evaluation formula:

[0315] P0=Y0

[0316] P=Y1×X1×K1+Y2×X2×K2+…+Yn×Xn×Kn

[0317] Calculate the value of the data asset.

[0318] Among them, Yn represents the contribution value of the nth single scene, Xn represents the number of the nth single scene, and Kn represents the correction coefficient of the nth single scene.

[0319] Formula for initial valuation of data assets:

[0320] Where n represents the available scenarios, and y represents the contribution in a single scenario; it can be used to evaluate the initial value of data assets, that is, the valuation of data assets in the first year.

[0321] The initial value of the scenario correction coefficient K1 corresponding to Y1 is 1. The ratio of the actual number of usage scenarios in the two years after the evaluation to the number of usage scenarios during the evaluation is K12, which serves as the scenario correction coefficient for the second year under the user entity; it is used in the data asset evaluation formula for the second year.

[0322] The ratio of the actual number of usage scenarios in the three years after the evaluation to the number of usage scenarios during the evaluation is K13, which serves as the scenario correction coefficient for the third year under the user entity; it is used in the data asset evaluation formula for the third year.

[0323] The ratio of the actual number of usage scenarios in the four years after the evaluation to the number of usage scenarios during the evaluation is K14, which serves as the scenario correction coefficient for the fourth year under the user entity; it is used in the fourth-year data asset evaluation formula.

[0324] The initial value of the scenario correction coefficient K2 corresponding to Y2 is 1. The ratio of the actual number of usage scenarios in the two years after the evaluation to the number of usage scenarios during the evaluation is K22, which serves as the scenario correction coefficient for the second year under the user entity; it is used in the data asset evaluation formula for the second year.

[0325] The ratio of the actual number of usage scenarios in the three years after the evaluation to the number of usage scenarios during the evaluation is K23, which serves as the scenario correction coefficient for the third year under the user entity; it is used in the data asset evaluation formula for the third year.

[0326] The ratio of the actual number of usage scenarios in the four years after the evaluation to the number of usage scenarios during the evaluation is K24, which serves as the scenario correction coefficient for the fourth year under the user entity; it is used in the fourth-year data asset evaluation formula.

[0327] The initial value of the scenario correction coefficient Kn corresponding to Yn is 1. The ratio of the actual number of usage scenarios in the two years after the evaluation to the number of usage scenarios during the evaluation is Kn2, which serves as the scenario correction coefficient for the second year under the user entity; it is used in the data asset evaluation formula for the second year.

[0328] The ratio of the actual number of usage scenarios in the three years after the evaluation to the number of usage scenarios during the evaluation is Kn3, which serves as the scenario correction coefficient for the third year under the user entity; it is used in the data asset evaluation formula for the third year.

[0329] The ratio of the actual number of usage scenarios in the four years after the evaluation to the number of usage scenarios during the evaluation is Kn4, which serves as the scenario correction coefficient for the fourth year under the user entity; it is used in the fourth-year data asset evaluation formula.

[0330] The second year evaluation formula is:

[0331] P=Y1×X1×K12+Y2×X2×K22+…+Yn×Xn×Kn2

[0332] The third year evaluation formula is:

[0333] P=Y1×X1×K13+Y2×X2×K23+…+Yn×Xn×Kn3

[0334] The fourth year evaluation formula is:

[0335] P=Y1×X1×K14+Y2×X2×K24+…+Yn×Xn×Kn4

[0336] Evaluation formula for the Nth year:

[0337] P=Y1×X1×K1n+Y2×X2×K2n+…+Yn×Xn×Knn

[0338] If the contribution is decreasing, you can recalculate the contribution and substitute it into the above formula:

[0339] The contribution of a single scenario in the second year is the difference between the value generated by the second use of the data asset and the value generated by the first use of the data asset, and so on.

[0340] The contribution of a single scenario in the third year is the difference between the final value generated by the third use of the data asset and the previous value generated by the second use of the data asset, and so on.

[0341] The contribution of a single scenario in the nth year is the difference between the value generated by the nth use of the data asset and the previous value generated by the n-1th use of the data asset, and so on.

[0342] Evaluation report generation module: used to generate evaluation reports and list the data scenarios used by the data asset, unused data scenarios, evaluation valuation, scenario correction coefficient, and number of scenarios.

[0343] The data asset evaluation method and system of the present invention have the following beneficial effects:

[0344] (1) This invention uses the number of scenarios and the degree of scenario contribution for evaluation, which reduces the requirements and evaluations of data quality by the cost method, benefit method, and market method, and effectively grasps the core of data asset evaluation.

[0345] (2) The present invention adopts the sum of the contributions of multiple scenarios, taking into account the practical problems of multi-scenario applications of a single data asset, and gives higher value to the data asset.

[0346] (3) The scenario coefficient correction in the present invention makes it possible to use multiple scenarios and to supplement the actual use of multiple scenarios, reducing the number of unrealizable scenarios in the evaluation, and is more in line with the requirement that the value of data asset evaluation depends on the ability of the user, providing an accurate and reliable basis for data asset evaluation.

[0347] (4) The present invention uses a neural network to perform scenario-based predictions on factors that actually affect usage scenarios, and trains the model's accuracy in the use of scenario correction coefficients.

[0348] (5) The present invention adopts a data asset and data scenario database, a scenario quantity database, and a scenario correction coefficient database to improve the accuracy of data asset evaluation.

[0349] (6) The present invention adopts the enterprise data scenario correction coefficient to form the industry scenario correction coefficient, which is conducive to the input of the initial value of the data scenario coefficient according to the industry in the future.

[0350] The above description of the present invention is in conjunction with the accompanying drawings. It is obvious that the implementation of the present invention is not limited to the above-mentioned method. As long as various improvements are made using the method concept and technical solution of the present invention, or the concept and technical solution of the present invention are directly applied to other occasions without improvement, they are all within the scope of protection of the present invention.

Claims

1. The scenario-based data asset assessment method includes the following steps: Step 1: Evaluate the scenarios in which data assets can be used; Step 2: Calculate the contribution of a single determined scenario; Step 3: Evaluate the number of single scenarios; Step 4: Calculate the scene correction coefficient; Step 5: Evaluate the value of data assets.

2. The scenario-based data asset evaluation method according to claim 1, characterized in that: Step 1 of evaluating the scenarios in which data assets can be used is performed by: S11. Enter the data asset name and search in the data scenario database; S12. The search results show the specific scenario name, the problem solved by the scenario, the number of scenarios, the industry name, and the industry name. The actual result is the number of known scenarios. S13. Determine specific scenarios in which the assessed entity can implement its capabilities.

3. The scenario-based data asset evaluation method according to claim 1, characterized in that: The method for establishing the data scenario database in step 1 is as follows: S21. Data asset classification; S22. Process the data scenarios by arranging, summarizing, and classifying them; S23. Establish the corresponding relationship between data assets and data scenarios; S24. Set up a dual search technology route based on data asset name and data scenario; S25. Auxiliary indicator search sets up national economic industry search.

4. The scenario-based data asset evaluation method according to claim 1, characterized in that: The method for establishing the data scenario contribution database in step 2 is as follows: S31. Classify data assets and establish different problem scenarios; S32. Collect and organize the values before use; S33. Calculate the contribution value of the data asset to the scenario after use; S34, calculating the value after use minus the value before use; S35. Set enterprise, industry, and sector attribute indicators to match scenario contributions.

5. The scenario-based data asset evaluation method according to claim 1, characterized in that: The method for establishing the scene correction coefficient database in step 4 is as follows: S51. Data collation, recording annual inputs: funds, personnel, market competition, business model, enterprise scale and other indicators; S52. Record the actual number of scenario-based uses in a single scenario per year; S53. Use neural network model training to obtain a data scenario quantity model for the next year; S54. Input data to calculate the scenario correction coefficient for the next year.

6. The scenario-based data asset evaluation method according to claim 1, characterized in that: The formula for evaluating the value of data assets in step 5 is: P=Y1×X1×K1+Y2×X2×K2+…+Yn×Xn×Kn Among them, Yn represents the contribution value of the nth single scene, Xn represents the number of the nth single scene, and Kn represents the correction coefficient of the nth single scene.

7. Scenario-based data asset assessment system, including the following systems: Data scenario database module: used to query scenarios for data assets, establish a correspondence between data asset reclassification and data scenarios (i.e., single specific issues), and try to grasp the range of known scenarios in which data assets may be used. It also establishes scenario innovation methods for unknown scenarios to generate new data scenarios for the data assets. Contribution calculation module: used to calculate the contribution of a single scenario to solving a problem; Scenario quantity database module: used for calculating the number of replicable usage scenarios of a single scenario in different life cycle years for multiple scenarios within the subject control cycle of data assets; Scenario correction coefficient database module: used for scenario correction coefficients in different industries; When establishing the industry scenario correction coefficient, factors such as enterprise size, personnel, financial strength, market share, number of enterprises, and industry development cycle are considered; At the same time, we also use the factors that affect the use of data scenarios within the enterprise to calculate the enterprise correction coefficients in different years and the industry data scenario correction coefficients; Evaluation module: used to evaluate the value of the data asset; Evaluation report generation module: used to generate evaluation reports and list the data scenarios used by the data asset, unused data scenarios, evaluation valuation, scenario correction coefficient, and number of scenarios.

Images