System contribution rate evaluation method based on exploratory simulation analysis

Abstract

The invention provides a system contribution rate evaluation method based on exploratory simulation analysis, which is used for objectively and accurately evaluating a system contribution rate and improving the reliability of an evaluation result. The method comprises the steps that exploration space data is generated based on a pre-established high-resolution simulation model, the high-resolution simulation model comprises an element model set and a strategy model set, and the element model set at least comprises physical elements, information elements and debate elements; constructing a plurality of key variables for debate subjects in the physical elements, the information elements, the debate elements and the support viewpoint elements, wherein the exploration space data comprises sample data generated based on the elements; according to the exploration space data, the confrontation efficiency is determined based on preset evaluation indexes and weights corresponding to the evaluation indexes, and the evaluation indexes comprise loss and achievements; and determining the contribution rate of the element system according to the confrontation efficiency.

Description

Technical Field

[0001] The present invention belongs to the technical field of simulation modeling, and particularly relates to a method for evaluating the contribution rate of a system based on exploratory simulation analysis. Background Art

[0002] In some typical group confrontation scenarios, such as large-scale group debate scenarios, it usually includes elements in different fields such as physics and information. The debaters are the main body of the debate system. Debate behaviors can effectively affect factors such as the positive emotions, negative emotions, and viewpoints of the debate main body, and play a crucial role in the success or failure of the participating parties. By studying the contribution rate of debate elements such as the debate main body and debate behaviors to the system of the participating parties, it can effectively support the system construction decision-making of the participating parties.

[0003] Traditional methods for evaluating the contribution rate of a system include methods such as analysis, statistical analysis, uncertainty reasoning, complex networks, and exploratory simulation analysis. Among them, the three methods of analysis, statistical analysis, and uncertainty reasoning are designed based on the idea of decomposition and reduction, and it is difficult to adapt to the analysis of system confrontation problems with microscopic large-scale uncertainty and macroscopic overall emergence, and they rely heavily on expert experience; in the system confrontation environment, although complex networks can describe the system characteristics from the perspective of macroscopic structure, they face problems such as difficult description of microscopic uncertainty and difficult selection of credible indicators, thus reducing the reliability of the evaluation results. Summary of the Invention

[0004] In order to objectively and accurately evaluate the contribution rate of a system and improve the reliability of the evaluation results of the contribution rate of a system, an embodiment of the present invention provides a method for evaluating the contribution rate of a system based on exploratory simulation analysis.

[0005] In an embodiment of the present invention, a method for evaluating the contribution rate of a system based on exploratory simulation analysis is provided, including:

[0006] Generating exploration space data based on a pre-established high-resolution simulation model, where the high-resolution simulation model includes an element model set and a strategy model set, and the element model set at least includes physical elements, information elements, and debate elements; constructing a number of key variables for the debate main body among the physical elements, information elements, debate elements, and supporting viewpoint elements respectively, and the exploration space data includes sample data generated based on each element;

[0007] According to the exploration space data, determining the confrontation effectiveness based on a pre-set evaluation index and its corresponding weight, and the evaluation index includes losses and achievements;

[0008] Determining the contribution rate of the element system according to the confrontation effectiveness.

[0009] In one implementation manner, the loss ratio of any element is determined according to the following formula:

[0010]

[0011] Among them: LOSS represents the loss ratio corresponding to this element, LOSSL i represents the loss quantity of the i-th element, FC i represents the importance degree of the i-th element, i represents the number of elements, LOSS i represents the number of the smallest identifiable units included in the i-th element.

[0012] In one implementation, the loss ratio of the physical element is determined according to the following formula:

[0013]

[0014] Among them, PFL represents the loss ratio corresponding to the physical element, represents the loss quantity of the i1-th physical element, represents the importance degree of the i1-th physical element, n1 represents the number of physical elements, represents the number of the smallest identifiable units included in the i1-th physical element.

[0015] In one implementation, the loss ratio of the information element is determined according to the following formula:

[0016]

[0017] Among them, IFL represents the loss ratio corresponding to the information element, represents the loss quantity of the i2-th information element, represents the importance degree of the i2-th information element, n2 represents the number of information elements, represents the number of the smallest identifiable units included in the i2-th information element.

[0018] In one implementation, the loss ratio of the debate element is determined according to the following formula:

[0019]

[0020] Among them, CFL represents the loss ratio corresponding to the debate element, represents the loss quantity of the i3-th debate element, represents the importance degree of the i3-th debate element, n3 represents the number of debate elements, represents the number of the smallest identifiable units included in the i3-th debate element.

[0021] In one implementation, the loss ratio of the supporting view is determined according to the following formula:

[0022]

[0023] Among them, SPOL represents the loss ratio corresponding to the support view element, SPO P represents the number of debate subjects with a support view before the confrontation, SPO A represents the number of confrontation subjects with a support view after the confrontation.

[0024] In one implementation, according to the confrontation effectiveness, determine the contribution rate of the element system:

[0025]

[0026] Among them, CRS represents the contribution rate of the element system, SE IC represents the confrontation effectiveness obtained when the participating party system includes this element, SE NIC represents the confrontation effectiveness obtained when the participating party system does not include this element.

[0027] In one implementation, set the system element decision variable, and the exploration space data is described in the following way:

[0028]

[0029] Among them, F(x k ) represents the decision variable value combination, x k represents the kth decision variable, which is used to describe the quantity of the kth element, m is the number of element types, and Z represents an integer.

[0030] In one implementation, determine the number of simulation runs according to the following formula:

[0031]

[0032] Among them, Cnt represents the number of simulation runs, r represents the number of decision variables, j represents the value quantity of x k , represents the calculation formula of the decision variable value combination.

[0033] In one implementation, the result is characterized by the loss ratio of the other party.

[0034] An evaluation method for system contribution rate based on exploratory simulation analysis provided by an embodiment of the present invention pre - establishes a high - resolution policy model, which includes an element model set and a strategy model set. Among them, the element model set includes at least physical elements, information elements, and debate elements; the strategy model set includes supporting view elements. The exploration space data generated based on this model covers the adversarial subject attribute data, adversarial subject behavior data, and behavior effect data. The evaluation and analysis of the adversarial effectiveness of the element system are carried out based on the generated exploration space data, making the evaluation results more objective and reasonable, thereby improving the reliability of the system efficiency evaluation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 FIG. is a schematic diagram of the implementation process of the system contribution rate evaluation method based on exploratory simulation analysis according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] Based on exploratory simulation analysis, the simulation method is used to simulate a large number of uncertain dynamic interactions existing microscopically in the adversarial system, and the effectiveness of the macroscopic system is analyzed using simulation data and statistical indicators. Based on the exploratory simulation analysis method, the present invention proposes a macroscopic evaluation index system, constructs simulation models of microscopic subjects, behaviors, and behavior effects, and applies the improved method to the problem of evaluating the system contribution rate of elements.

[0037] As Figure 1 shown, it is a schematic diagram of the implementation process of the system contribution rate evaluation method based on exploratory simulation analysis provided by an embodiment of the present invention, including the following steps:

[0038] S11. Generate exploration space data based on a pre - established high - resolution simulation model.

[0039] In the present invention, for the group debate confrontation scenario, a high - resolution simulation model is selected to describe the element model set and strategy model set of both sides of the confrontation, and the formal description is:

[0040] HRMset = {RBFset, RBAset}

[0041] Among them, HRMset represents the high - resolution model set, RBFset represents the element model set of both sides of the debate, and RBAset represents the strategy model set of both sides.

[0042] The high - resolution elements include 3 categories: physical, information, and debate, and the formal description is:

[0043] RBFset = {PhEl, InfEl, PubEl}

[0044] Among them, PhEl represents the set of physical elements, InfEl represents the set of information elements, and PubEl represents the set of debate elements.

[0045] For the debate subjects in physical elements, information elements, and debate elements, three key variables of positive emotion, negative emotion, and opinion need to be constructed, and the formal description is as follows:

[0046]

[0047] Among them, PFactset represents the set of attributes of the debate subject, Mor represents positive emotion information, Emo represents negative emotion information, and Atd represents opinion-related information.

[0048] Set the decision variables of the system elements, and the formal description is as follows:

[0049]

[0050] Among them, F(x i ) represents the combination of decision variable values, x i represents the i-th decision variable, which is used to describe the quantity of the i-th element, n is the number of element types, and Z represents an integer.

[0051] In the exploratory simulation analysis, the calculation formula for the number of simulation runs is:

[0052]

[0053] Among them, Cnt represents the number of simulation runs, n represents the number of decision variables, x i represents the i-th decision variable, j represents the number of values of x i , represents the calculation formula for the combination of decision variable values.

[0054] According to the decision variables and the number of runs defined above, run the constructed microscopic high-resolution simulation model to obtain the corresponding running results, and each running result corresponds to a sample data.

[0055] S12. According to the generated exploration space data, determine the confrontation effectiveness based on the pre-set evaluation indicators and their corresponding weights.

[0056] In the embodiments of the present invention, the confrontation effectiveness evaluation index system is defined as follows:

[0057] The effectiveness of the participating party system includes two important indicators: the losses brought by the opponent to oneself and the achievements obtained by oneself, and the formal description is as follows:

[0058] SE = {Loss, Eff}

[0059] Among them, SE represents the effectiveness of the participating party system, Loss represents the losses brought by the opponent to one's own side, and Eff represents the achievements obtained by one's own side.

[0060] The loss Loss includes four indicators: the loss ratio of physical elements, the loss ratio of information elements, the loss ratio of debate elements, and the loss ratio of supported viewpoints, and is formally described as:

[0061] Loss = {PFL, IFL, CFL, SPOL}

[0062] Among them, PFL represents the loss ratio of physical elements, IFL represents the loss ratio of information elements, CFL represents the loss ratio of debate elements, and SPOL represents the loss ratio of supported viewpoints.

[0063] In specific implementation, for each element, the loss ratio of this element can be calculated according to the following formula:

[0064]

[0065] Among them: LOSS represents the loss ratio corresponding to this element, LOSSL i represents the loss quantity of the i-th element, FC i represents the importance degree of the i-th element, i represents the number of elements, LOSS i represents the number of the smallest identifiable units included in the i-th element.

[0066] Specifically, the calculation formula for the loss ratio of physical elements is:

[0067]

[0068] Among them, PFL represents the loss ratio corresponding to the physical elements, represents the loss quantity of the i1-th physical element, represents the importance degree of the i1-th physical element, n1 represents the number of physical elements, represents the number of the smallest identifiable units included in the i1-th physical element.

[0069] The calculation formula for the loss ratio of information elements is:

[0070]

[0071] Among them, IFL represents the loss ratio corresponding to the information elements, represents the loss quantity of the i2-th information element, represents the importance degree of the i2-th information element, n2 represents the number of information elements, represents the number of the smallest identifiable units included in the i2-th information element.

[0072] The calculation formula for the loss ratio of debate elements is as follows:

[0073]

[0074] Among them, CFL represents the loss ratio corresponding to the debate element, represents the loss quantity of the i3-th debate element, represents the importance degree of the i3-th debate element, and n3 represents the number of debate elements, represents the number of the smallest identifiable units included in the i3-th debate element.

[0075] The calculation formula for the loss ratio of supporting viewpoints is as follows:

[0076]

[0077] Among them, SPOL represents the loss ratio corresponding to the supporting viewpoint element, SPO P represents the number of debate subjects holding supporting viewpoints before the confrontation, SPO A represents the number of confrontation subjects holding supporting viewpoints after the confrontation.

[0078] The achievement Eff includes 4 indicators: the loss of the opponent's physical elements, the loss of the opponent's information elements, the loss of debate elements, and the change rate of opposing viewpoints. The formal description is as follows:

[0079] Eff = {EPFL, EIFL, ECFL, OPOL}

[0080] It should be noted that the achievement of one's own side can be characterized by the loss ratio of the other side. Therefore, in specific implementation, EPFL represents the loss ratio of the other side's physical elements, and the calculation formula of this indicator refers to the above-mentioned calculation formula for the loss ratio of physical elements. EIFL represents the loss ratio of the other side's information elements, and the calculation formula of this indicator refers to the above-mentioned calculation formula for the loss ratio of information elements. ECFL represents the loss ratio of the other side's debate elements, and the calculation formula of the indicator refers to the above-mentioned calculation formula for the loss ratio of debate elements. OPOL represents the loss ratio of the other side's supporting viewpoints, and the calculation formula of this indicator refers to the above-mentioned calculation formula for the loss ratio of supporting viewpoints.

[0081] Regarding the weights of each evaluation indicator, in specific implementation, they can be set according to experience. In the embodiments of the present invention, an indicator weight questionnaire can be made, and the expert scoring method can be used to obtain the weights of each indicator. The formal description is as follows:

[0082]

[0083] Among them, represents the weight of the j-th indicator, represents the assignment of the i-th expert to the weight of the j-th indicator, and n represents the number of experts.

[0084] Determine the corresponding confrontation effectiveness according to the index values of each index obtained from the above calculations and their corresponding weights.

[0085] S13. Determine the contribution rate of the element system according to the determined confrontation effectiveness.

[0086] It should be noted that in specific implementation, the higher the confrontation effectiveness, the higher the system contribution.

[0087] To better understand the present invention, the implementation process of the present invention will be described below in conjunction with specific embodiments. The system contribution rate evaluation method based on exploratory simulation analysis provided by the embodiments of the present invention is divided into three steps: exploratory space design, multi-resolution modeling, and simulation operation and evaluation. The specific steps are as follows:

[0088] S1. Exploratory space design

[0089] Set the element decision variable, which is formally described as:

[0090]

[0091] where x i represents the i-th decision variable, used to describe the quantity of the i-th element, n is the number of element types, and Z represents an integer.

[0092] In exploratory simulation analysis, the calculation formula for the number of simulation runs is:

[0093]

[0094] where Cnt represents the number of simulation runs, n represents the number of decision variables, x i represents the i-th decision variable, j represents the number of values of x i , represents the calculation formula for the decision variable value combination.

[0095] S2. Multi-resolution modeling

[0096] (1) Low-resolution system contribution rate calculation model

[0097] The formal calculation formula for the system contribution rate is:

[0098]

[0099] where CRS represents the system contribution rate, SE IC represents the confrontation effectiveness obtained when the participating party's system includes debate elements, and SE NIC represents the confrontation effectiveness obtained when the participating party's system does not include debate elements.

[0100] (2) Confrontation effectiveness evaluation index system

[0101] The effectiveness of the participant system includes two important indicators: the losses brought by the opponent to oneself and the achievements obtained by oneself. It is formally described as:

[0102] SE = {Loss, Eff}

[0103] Among them, SE represents the effectiveness of the participant system, Loss represents the losses brought by the opponent to oneself, and Eff represents the achievements obtained by oneself.

[0104] The battle losses Loss includes four indicators: the loss ratio of physical elements, the loss ratio of information elements, the loss ratio of debate elements, and the loss ratio of view support. It is formally described as:

[0105] Loss = {PFL, IFL, CFL, SPOL}

[0106] Among them, PFL represents the loss ratio of physical elements, IFL represents the loss ratio of information elements, CFL represents the loss ratio of debate elements, and SPOL represents the loss ratio of view support.

[0107] The calculation formula for the loss ratio of physical elements is:

[0108]

[0109] Among them, PFL i represents the loss quantity of the i-th physical element, FC i represents the importance degree of the i-th physical element, i represents the quantity of physical elements, PF i represents the quantity of the smallest identifiable units included in the i-th physical element.

[0110] The calculation formula for the loss ratio of information elements is:

[0111]

[0112] Among them, IFL i represents the loss quantity of the i-th information element, FC i represents the importance degree of the i-th information element, i represents the quantity of information elements, IF i represents the quantity of the smallest identifiable units included in the i-th information element.

[0113] The calculation formula for the loss ratio of debate elements is:

[0114]

[0115] Among them, CFL i represents the loss quantity of the i-th debate element, FC iIndicates the importance level of the i-th debate element, where i represents the number of debate elements, IF i Indicates the number of the smallest recognizable units included in the i-th debate element.

[0116] The calculation formula for the loss ratio of view support is:

[0117]

[0118] Among them, SPO P Indicates the number of people who held a supportive view before the confrontation, and SPO A Indicates the number of people who held a supportive view after the confrontation.

[0119] The battle result Eff includes 4 indicators: the loss of the opponent's physical elements, the loss of the opponent's information elements, the loss of debate elements, and the change rate of view opposition. The formal description is:

[0120] Eff = {EPFL, EIFL, ECFL, OPOL}

[0121] Among them, EPFL indicates the loss ratio of the opponent's physical elements. The calculation formula for this indicator refers to the calculation formula for the loss ratio of physical elements. EIFL indicates the loss ratio of the opponent's information elements. The calculation formula for this indicator refers to the calculation formula for the loss ratio of information elements. ECFL indicates the loss ratio of debate elements. The calculation formula for the indicator refers to the calculation formula for the loss ratio of debate elements. OPOL indicates the loss ratio of the opponent's view support. The calculation formula for this indicator refers to the calculation formula for the loss ratio of view support.

[0122] (3) Assignment of indicator weights

[0123] Make an indicator weight questionnaire, and use the expert scoring method to obtain the weights of each indicator. The formal description is:

[0124]

[0125] Among them, Indicates the weight of the j-th indicator, Indicates the assignment of the weight of the j-th indicator by the i-th expert, and n represents the number of experts.

[0126] (4) High-resolution simulation model

[0127] Select a high-resolution simulation model to describe the element model set and strategy model set of both sides of the confrontation. The formal description is:

[0128] HRMset = {RBFset, RBAset}

[0129] Among them, HRMset represents the high-resolution model set, RBFset represents the bilateral factor model set, and RBAset represents the bilateral strategy model set. Among them, three key variables, namely positive emotion, negative emotion, and opinion, are selected in the factor model to describe the impact of the debate on people.

[0130] S3. Simulation operation and evaluation

[0131] According to the decision variables and the number of simulation runs in S1, run the microscopic high-resolution simulation model constructed in S2, and use the macroscopic low-resolution evaluation model in S2 to statistically analyze the operation data of the high-resolution simulation model to form a system contribution rate score.

[0132] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A method for evaluating the contribution rate of a system based on exploratory simulation analysis, characterized in that Including: Generating exploration space data based on a pre-established high-resolution simulation model, the high-resolution simulation model including an element model set and a policy model set, the element model set at least including physical elements, information elements, and debate elements; constructing a number of key variables for the debate subjects in the physical elements, information elements, debate elements, and supporting view elements respectively, the exploration space data including sample data generated based on each element; According to the exploration space data, determining the confrontation effectiveness based on a pre-set evaluation index and its corresponding weight, the evaluation index including losses and achievements; Determining the contribution rate of the element system according to the confrontation effectiveness.

2. The method according to claim 1, wherein Determining the loss ratio of any element according to the following formula: Where: LOSS represents the loss ratio corresponding to the element, LOSSL i represents the loss quantity of the i-th element, FC i represents the importance level of the i-th element, i represents the number of elements, LOSS i represents the number of the minimum recognizable units included in the i-th element.

3. The method according to claim 2, wherein Determining the loss ratio of physical elements according to the following formula: where FPL represents the loss ratio corresponding to the physical element, represents the loss quantity of the i1-th physical element, represents the importance degree of the i1-th physical element, and n1 represents the number of physical elements, represents the number of the minimum recognizable units included in the i1-th physical element.

4. The method according to claim 2, wherein Determining the loss ratio of information elements according to the following formula: Among them, IFL represents the loss ratio corresponding to the information element, represents the loss quantity of the i2-th information element, represents the importance degree of the i2-th information element, and n2 represents the number of information elements, represents the number of the smallest recognizable units included in the i2-th information element.

5. The method according to claim 2, characterized in that, Determining the loss ratio of debate elements according to the following formula: Among them, CFL represents the loss ratio corresponding to the debate element, represents the loss quantity of the i3-th debate element, represents the importance degree of the i3-th debate element, and n3 represents the number of debate elements, represents the minimum recognizable unit quantity contained in the i3-th debate element.

6. The method according to claim 2, wherein Determining the loss ratio of the supporting view according to the following formula: Among them, SPOL represents the loss ratio corresponding to the supporting opinion element, and SPO P represents the number of debate subjects holding a supporting opinion before the confrontation, and SPO A represents the number of confrontation subjects holding a supporting opinion after the confrontation.

7. The method according to claim 1, characterized in that Determining the contribution rate of the element system according to the confrontation effectiveness: Among them, CRS represents the contribution rate of the element system, and SE IC represents the confrontation effectiveness obtained when the participating party system includes this element, and SE NIC represents the confrontation effectiveness obtained when the participating party system does not include this element.

8. The method according to claim 1, characterized in that, Setting system element decision variables, the exploration space data is described in the following way: Among them, F(x k ) represents the combination of decision variable values, x k represents the k-th decision variable, which is used to describe the quantity of the k-th element. m is the number of element types, and Z represents an integer.

9. The method according to claim 8, characterized in that, Determining the number of simulation runs according to the following formula: Among them, Cnt represents the number of simulation runs, r represents the number of decision variables, and j represents the number of value-taking of x k The number of value-taking, represents the calculation formula for the value-taking combination of decision variables.

10. The method according to any one of claims 1-9, characterized in that The achievement is characterized by the opponent's loss ratio.

Images