Emergency mission planning scheme evaluation method and system based on personalized feedback mechanism

Abstract

The present invention provides an emergency task planning scheme evaluation method, system, storage medium and electronic device based on a personalized feedback mechanism, and relates to the technical field of emergency task planning scheme evaluation. The present invention obtains the social network relationship between each evaluator, and the trust value between each evaluator in the social network relationship, and is no longer limited to the premise that the evaluators are independent of each other; when the consensus of a certain evaluator does not reach a predefined consensus threshold, a personalized feedback mechanism is used to adjust the personal decision preference matrix of the evaluator who has not reached a consensus, and obtain the adjusted group decision preference matrix. The personalized feedback mechanism based on the Stackelberg game architecture not only ensures that the group consensus level required for group decision-making is achieved under the condition of minimum adjustment cost and maximum harmony between evaluators, but also takes into account the personal characteristics of the evaluators to achieve group diversity, and avoids the situation of reaching consensus for the sake of reaching consensus.

Description

Technical Field

[0001] The present invention relates to the technical field of emergency task planning scheme evaluation, and in particular to an emergency task planning scheme evaluation method, system, storage medium and electronic equipment based on a personalized feedback mechanism. Background Art

[0002] China has a vast territory and a complex geographical environment. Typhoons, floods, forest fires and other emergencies occur frequently. Remote sensing technology, as a "revolutionary extension of human visual perception", has the advantages of being comprehensive, fast and intuitive, and is not limited by time and region. Therefore, remote sensing technology provides a good way to quickly obtain real-time information about disaster areas. However, the remote sensing satellite observation system is large and complex, with high risks and high investments. In order to reduce the manpower, material and financial resources required to operate such a system, it is necessary to select an optimal planning scheme to ensure that satellite resources are fully utilized to obtain as much disaster area information as possible in the shortest time. In view of the complexity and uncertainty of the planning scheme selection process, the evaluation problem of emergency mission planning schemes is usually regarded as an emergency decision-making problem.

[0003] Generally speaking, emergency decision-making has two distinct characteristics: first, decisions must be made in a short period of time based on partial or incomplete information, especially in the early stages of an emergency; second, the evaluation of emergency task planning options may have potentially serious consequences; in other words, an inaccurate decision may lead to catastrophic losses. Since emergency decision-making involves a wide range of issues, it often requires the participation of evaluators from different professional backgrounds and has the complex characteristics of a large group, which can be called large-group emergency evaluation (LGEE). At present, LGEE has received increasing attention in the field of decision science, and many group decision-making models have been proposed.

[0004] Existing evaluation methods for emergency task planning programs are mainly based on the premise that evaluators are independent of each other. However, with the development of social networks, this premise has become difficult to meet, which directly affects the quality of evaluation. Summary of the invention

[0005] 1. Technical issues to be solved

[0006] In view of the deficiencies in the prior art, the present invention provides an emergency task planning scheme evaluation method, system, storage medium and electronic device based on a personalized feedback mechanism, which solves the technical problem of low evaluation quality.

[0007] (II) Technical solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] An emergency task planning scheme evaluation method based on personalized feedback mechanism, comprising:

[0010] According to a number of pre-given emergency task planning schemes, the evaluation information of each evaluator corresponding to each evaluation indicator under each emergency task planning scheme is obtained, and the personal decision preference matrix of each evaluator is determined;

[0011] Obtaining the social network relationships between the evaluators and the trust values ​​between the evaluators in the social network relationships;

[0012] According to the trust value in the social network relationship, the weight of each evaluator is obtained; and the group decision preference matrix is ​​obtained by aggregating the individual decision preference matrices of each evaluator through the weights;

[0013] Calculate the consensus of each evaluator. When the consensus of a certain evaluator does not reach the pre-defined consensus threshold, use the personalized feedback mechanism to adjust the individual decision preference matrix of the evaluator who has not reached a consensus, and obtain the adjusted group decision preference matrix;

[0014] According to the adjusted group decision preference matrix, a recommended ranking of several emergency task planning schemes is obtained to determine the optimal emergency task planning scheme.

[0015] Preferably, determining the individual decision preference matrix of each evaluator includes:

[0016] Define n emergency task planning schemes given in advance with X = {x1, x2, …, x n}, each planning scheme corresponds to a predetermined m evaluation index, which is represented by C = {C1, C2, …, C m} indicates that the pre-given first language item set S = {s t |t=0,1,…,τ}, t represents the subscript of the language item, and the total number of language items in the first language item set is τ+1;

[0017] is assessed by s Given an individual decision preference matrix, a set of possible language items is used as the expression form; Represents the evaluator s Given an emergency mission planning solution x i For the evaluation index C j preference information, Representation elements The number of language items in , Represents a language item, In the planning scheme x iFor the evaluation index C j Chinese Language The corresponding probability;

[0018] When there is When the decision preference matrix given by all evaluators is standardized so that the sum of the probabilities corresponding to all language items is equal to 1, the standardized form of preference information is

[0019] in in, It means that the deviation of the probability and corresponding to 1 and all language items is divided equally into (1+τ) parts, and Λ represents the preference information The set of all language item subscripts in the first language item set S, Γ = {0, 1, ..., τ}; ρ represents all the items in the first language item set S but not in the preference information The set of all language item subscripts in .

[0020] Preferably, the obtaining of the trust value between each evaluator in the social network relationship includes:

[0021] T=(t sr ) K×K It represents the trust value between each evaluator in a given social network relationship, and adopts a set of possible degree language items as the expression form; K represents the total number of evaluators in the evaluation process, is a set of possible degree linguistic items, representing the evaluator e s For r Trust value, H = {h σ |σ=0,1,…,χ} is a pre-given set of second language items, σ represents the subscript of the second language item set H, and the total number of language items in the second language item set is χ+1. Represents a language item, Represents the evaluator s For r The language item in the trust value The corresponding probability;

[0022] When there is When , the trust value between each evaluator is standardized. express, The calculation formula is as follows:

[0023]

[0024] in, It means that the deviation of the probability sum corresponding to 1 and all language items is divided into (1+χ) parts evenly, and Λ1 represents the trust value tsr The set of all language item subscripts in , Γ1={0,1,…,χ}; ρ1 represents all items in the second language set H but not in the trust value t sr The set of all language item subscripts in .

[0025] Preferably, if the social network relationship is incomplete, the Archimedean t-norm probability language trust propagation operator P is introduced pl To estimate the indirect trust value between unknown evaluators in the social network relationship, including:

[0026]

[0027] in, represents the standardized trust value of the ith edge in the trust propagation chain built based on social network relationships, and σ represents the ith trust value Medium language item σ The subscript value of (q σ ) N,i Trust value Medium language item σ The corresponding probability; g -1 (β) = h βσ It means converting the value β∈[0,1] into a language item in the second language item set H.

[0028] Preferably, obtaining the weight of each evaluator according to the trust value in the social network relationship includes:

[0029] According to the trust value in the social network relationship, the expected degree of trust value between any two evaluators is:

[0030] Calculate the evaluator s Trust level:

[0031] Calculate the weight of each evaluator:

[0032]

[0033] Among them, ε is a sort, PT ε(l) represents the set {RT1,RT2,…,RT K} the lth largest element in; represents the set {RT1,RT2,…,RT K The sum of the trustworthiness of the first s elements in}, and It represents a fuzzy linguistic quantifier;

[0034] The individual decision preference matrices of each evaluator are aggregated by weight to obtain a group decision preference matrix, including:

[0035] The weighted average operator is used to aggregate the decision preference matrices of all evaluators to obtain the group decision preference matrix The calculation formula is:

[0036]

[0037] Among them, W A represents the weighted average operator;

[0038] Represents the evaluator s The normalized form of the given decision preference matrix.

[0039] Preferably, determining that the consensus of a certain evaluator does not reach a predefined consensus threshold includes:

[0040] (1) Determine the set of evaluators who have not reached the consensus level: ECH = {s|CI s <γ},

[0041] in, Represents the evaluator s The consensus degree of ;γ represents the consensus threshold;

[0042] (2) Determine the set of solutions that do not reach the consensus level corresponding to the evaluators who do not reach the consensus level:

[0043] in, Represents the evaluator s Corresponding emergency task planning plan x i degree of consensus;

[0044] (3) Determine the set of elements that do not reach the consensus level, in, Represents the evaluator s Corresponding emergency task planning plan x i For the evaluation index C j degree of consensus;

[0045] Preferably, the personalized feedback mechanism is used to adjust the personal decision preference matrix of the evaluators who have not reached a consensus, specifically based on the Stackelberg game model to represent the interactive behavior between the host and the evaluator in the feedback adjustment process, including:

[0046] Minimize the host's adjustment costs:

[0047] Determine personalized feedback parameters for all inconsistent elements based on the collective APS Use this parameter to generate recommended evaluation information The calculation formula is:

[0048]

[0049] in, Represents the evaluator s In emergency mission planning program x i The corresponding evaluation index C j The personalized feedback parameters below.

[0050] The adjustment cost considered by the moderator is calculated by measuring the sum of the deviations between the initial preferences of all evaluators and the adjusted preference information, i.e.

[0051]

[0052] in, represents the adjusted preference information,

[0053] Maximize the satisfaction of each evaluator:

[0054] The adjustment range based on the evaluation information, i.e. Determine the specific value of the adjusted personal preference information, and the satisfaction calculation formula for each evaluator is:

[0055]

[0056] in, Denote any two evaluators e s and e r The deviation of the adjusted decision matrix, w sr Denote any two evaluators e s and e r The relative weight between

[0057] Preferably, obtaining the adjusted group decision preference matrix includes:

[0058] The WA operator is used to calculate the adjusted group decision preference matrix:

[0059] The calculation formula is:

[0060]

[0061] in, Represents the evaluator s The standardized form of the adjusted decision preference matrix.

[0062] Preferably, the determining of the optimal emergency task planning scheme specifically includes:

[0063] Aggregate the evaluation information of all evaluation indicators corresponding to each emergency task planning scheme to obtain the evaluation information of n observation schemes The calculation formula is as follows:

[0064]

[0065] Where ω=(ω1,ω2,…,ω m ) represents the weights corresponding to the pre-given m evaluation indicators;

[0066] Calculate the expected degree corresponding to the evaluation information of each emergency task planning scheme Compare the expected degree of each emergency task planning scheme, and select the observation scheme with the largest expected degree as the optimal emergency task planning scheme; the calculation formula of the expected degree is as follows:

[0067]

[0068] An emergency task planning scheme evaluation system based on personalized feedback mechanism, comprising:

[0069] The personal evaluation module is used to obtain the evaluation information of each evaluator corresponding to each evaluation indicator under each emergency task planning scheme according to a number of pre-given emergency task planning schemes, and determine the personal decision preference matrix of each evaluator;

[0070] A trust value acquisition module, used to acquire the social network relationship between each evaluator, and the trust value between each evaluator in the social network relationship;

[0071] A group decision module is used to obtain the weight of each evaluator according to the trust value in the social network relationship; and to obtain the group decision preference matrix by aggregating the individual decision preference matrices of each evaluator through the weights;

[0072] The decision adjustment module is used to calculate the consensus of each evaluator. When the consensus of a certain evaluator does not reach the pre-defined consensus threshold, a personalized feedback mechanism is used to adjust the individual decision preference matrix of the evaluator who has not reached a consensus, and obtain the adjusted group decision preference matrix;

[0073] The scheme recommendation module is used to obtain the recommended ranking of several emergency task planning schemes according to the adjusted group decision preference matrix, and determine the optimal emergency task planning scheme.

[0074] A storage medium stores a computer program for emergency mission planning scheme evaluation based on a personalized feedback mechanism, wherein the computer program enables a computer to execute the emergency mission planning scheme evaluation method as described above.

[0075] An electronic device, comprising:

[0076] one or more processors;

[0077] Memory; and

[0078] One or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, the programs including instructions for executing the emergency mission planning solution evaluation method as described above.

[0079] (III) Beneficial effects

[0080] The present invention provides an emergency task planning scheme evaluation method, system, storage medium and electronic device based on a personalized feedback mechanism. Compared with the prior art, it has the following beneficial effects:

[0081] The present invention includes obtaining evaluation information of each evaluator corresponding to each evaluation indicator under each emergency task planning scheme according to a plurality of pre-given emergency task planning schemes, and determining the personal decision preference matrix of each evaluator; obtaining the social network relationship between each evaluator, and the trust value between each evaluator in the social network relationship; obtaining the weight of each evaluator according to the trust value in the social network relationship; and aggregating the personal decision preference matrix of each evaluator by weight to obtain a group decision preference matrix; calculating the consensus of each evaluator, and when the consensus of a certain evaluator does not reach a pre-defined consensus threshold, using a personalized feedback mechanism to adjust the personal decision preference matrix of the evaluator who has not reached a consensus, and obtaining the adjusted group decision preference matrix; according to the adjusted group decision preference matrix, obtaining the recommended ranking of the plurality of emergency task planning schemes, and determining the optimal emergency task planning scheme.

[0082] On the one hand, it is no longer limited to the premise that evaluators are independent of each other, and the social network relationship and trust value between each evaluator are taken into consideration; on the other hand, a personalized feedback mechanism based on the Stackelberg game framework is proposed, which not only ensures the minimum adjustment cost and the maximum harmony between evaluators to achieve the level of group consensus required for group decision-making, but also takes into account the personal characteristics of the evaluators to achieve group diversity and avoid the situation of reaching consensus for the sake of reaching consensus. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0084] Figure 1 A flowchart of an emergency task planning scheme evaluation method based on a personalized feedback mechanism provided by an embodiment of the present invention;

[0085] Figure 2 A schematic diagram of a social network relationship between evaluators provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0086] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0087] The embodiments of the present application solve the technical problem of low evaluation quality by providing an emergency task planning scheme evaluation method, system, storage medium and electronic device based on a personalized feedback mechanism.

[0088] The technical solution in the embodiment of the present application is to solve the above technical problems, and the overall idea is as follows:

[0089] In practice, group evaluation is often conducted within the framework of social networks, and individual opinions are often supported by relatives, friends or like-minded people. And related research shows that it is more realistic to use language information to express the trust relationship between evaluators in social networks, but due to the limitations of cognitive ability, it is difficult for evaluators to provide their own complete probability distribution information when providing relevant probability information for each language term. In order to fill the gap between the trust evaluation model and actual needs, the embodiment of the present invention applies a probabilistic language trust function to describe the trust relationship between evaluators, which can more flexibly and accurately describe the trust relationship between groups.

[0090] In addition, in LGEE, the evaluators involved in decision-making come from different fields and have different interests, which means that reaching group consensus in LGEE takes more time and cost than reaching group consensus in general decision-making problems. Therefore, the consensus of LGEE requires a capable moderator to invest time and resources to persuade the evaluators to adjust their opinions to an acceptable level. However, the existing evaluation methods are often studied from the perspective of the moderator or the evaluator, ignoring the interaction mechanism between the moderator and the evaluator during the evaluation process. This defect may prevent the evaluators who have not reached a consensus from obtaining satisfactory adjustment suggestions during the consensus adjustment process, forcing them to reach the center of the group in a homogeneous manner, which ultimately affects the evaluation quality.

[0091] The Stackelberg game is a two-stage complete information dynamic game. Its main idea is that both parties choose their own strategies based on the other party's possible strategies to maximize their own interests under the other party's strategy, thereby achieving Nash equilibrium.

[0092] The embodiment of the present invention is based on the framework of Stackelberg game and develops an emergency task planning scheme evaluation method with a personalized feedback mechanism that considers maximum harmony and minimum adjustment cost in a social network environment. Different from the existing evaluation method that repeatedly modifies the evaluation information, the proposed evaluation method can meet the consensus threshold between evaluators through one round of iteration based on the optimization theory, thus shortening the time spent on large-scale group decision-making.

[0093] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0094] Example:

[0095] like Figure 1 As shown, an embodiment of the present invention provides an emergency task planning scheme evaluation method based on a personalized feedback mechanism, which is characterized by comprising:

[0096] S1. According to a number of pre-given emergency task planning schemes, obtain the evaluation information of each evaluator corresponding to each evaluation indicator under each emergency task planning scheme, and determine the personal decision preference matrix of each evaluator;

[0097] S2, obtaining the social network relationship between each evaluator, and the trust value between each evaluator in the social network relationship;

[0098] S3, obtaining the weight of each evaluator according to the trust value in the social network relationship; and obtaining the group decision preference matrix by aggregating the individual decision preference matrices of each evaluator through the weights;

[0099] S4. Calculate the consensus of each evaluator. When the consensus of a certain evaluator does not reach the pre-defined consensus threshold, use a personalized feedback mechanism to adjust the individual decision preference matrix of the evaluator who has not reached a consensus, and obtain the adjusted group decision preference matrix;

[0100] S5. According to the adjusted group decision preference matrix, obtain the recommended ranking of several emergency task planning schemes and determine the optimal emergency task planning scheme.

[0101] The advantages of the embodiments of the present invention include at least: on the one hand, it is no longer limited to the premise that the evaluators are independent of each other, and the social network relationship and trust value between each evaluator are taken into account; on the other hand, a personalized feedback mechanism based on the Stackelberg game architecture is proposed, which not only ensures that the level of group consensus required for group decision-making is achieved with the minimum adjustment cost and the maximum harmony between evaluators, but also takes into account the personal characteristics of the evaluators to achieve group diversity, avoiding the situation of reaching consensus for the sake of reaching consensus.

[0102] The following will introduce each step of the above technical solution in detail with specific content:

[0103] In step S1, according to a number of pre-given emergency task planning schemes, the evaluation information of each evaluator corresponding to each evaluation indicator under each emergency task planning scheme is obtained, and the personal decision preference matrix of each evaluator is determined.

[0104] Define n emergency task planning schemes given in advance with X = {x1, x2, …, x n}, each planning scheme corresponds to a predetermined m evaluation index, which is represented by C = {C1, C2, …, C m} indicates that the pre-given first language item set S = {s t |t=0,1,…,τ}, t represents the subscript of the language item, and the total number of language items in the first language item set is τ+1;

[0105] is assessed by s Given an individual decision preference matrix, a set of possible language items is used as the expression form; Represents the evaluator s Given an emergency mission planning solution x i For the evaluation index C j preference information, Representation elements The number of language items in , Represents a language item, In the planning scheme x i For the evaluation index C jChinese Language The corresponding probability;

[0106] When there is When the decision preference matrix given by all evaluators is standardized so that the sum of the probabilities corresponding to all language items is equal to 1, the standardized form of preference information is

[0107] in in, It means that the deviation of the probability and corresponding to 1 and all language items is divided equally into (1+τ) parts, and Λ represents the preference information The set of all language item subscripts in the first language item set S, Γ = {0, 1, ..., τ}; ρ represents all the items in the first language item set S but not in the preference information The set of all language item subscripts in .

[0108] In step S2, the social network relationships between the evaluators and the trust values ​​between the evaluators in the social network relationships are obtained.

[0109] The social network relationships between evaluators are as follows: Figure 2 As shown in , a social network is a social structure consisting of a set of nodes E and a set of edges L, where nodes E represent decision makers and directed edges L represent trust relationships. Figure 2 The pointer from e5 to e7 indicates that e5 trusts e7.

[0110] T=(t sr ) K×K It represents the trust value between each evaluator in a given social network relationship, and adopts a set of possible degree language items as the expression form; K represents the total number of evaluators in the evaluation process, is a set of possible degree linguistic items, representing the evaluator e s For r Trust value, H = {h σ |σ=0,1,…,χ} is a pre-given set of second language items, σ represents the subscript of the second language item set H, and the total number of language items in the second language item set is χ+1. Represents a language item, Represents the evaluator s For r The language item in the trust value The corresponding probability;

[0111] When there is When , the trust value between each evaluator is standardized. express, The calculation formula is as follows:

[0112]

[0113] in, It means that the deviation of the probability sum corresponding to 1 and all language items is divided into (1+χ) parts evenly, and Λ1 represents the trust value t sr The set of all language item subscripts in , Γ1={0,1,…,χ}; ρ1 represents all items in the second language set H but not in the trust value t sr The set of all language item subscripts in .

[0114] In step S3, the weight of each evaluator is obtained according to the trust value in the social network relationship; and the personal decision preference matrix of each evaluator is aggregated by weight to obtain the group decision preference matrix.

[0115] [1] Design a trust propagation operator to construct a complete trust relationship in social networks.

[0116] In real life, trust relationships in social networks are usually incomplete, that is, some evaluators may not be able to directly provide trust values ​​to unknown evaluators. Figure 2 In the example, there is no direct trust value between evaluators e1 and e3, but this does not mean that there is no trust relationship between evaluators e1 and e3. They can communicate through the trust propagation chain e1→e2→e 14 →e 11 →e3 to spread trust.

[0117] Therefore, based on social networks, we can build a trust chain to spread trust by establishing a trust relationship between a trusted third-party partner and an unknown evaluator. We can call this trust relationship indirect trust. If there are multiple trust propagation chains between two unknown evaluators, we can choose the shortest trust propagation chain to spread trust. However, the propagation of the trust propagation chain requires a trust propagation operator to achieve it. The trust propagation operator can be used to calculate the trust value of any two unknown evaluators based on the trust propagation chain, so as to determine whether the two unknown evaluators can trust each other.

[0118] In this step, if the social network relationship is incomplete, the Archimedean t-norm probability language trust propagation operator P is introduced. pl To estimate the indirect trust value between unknown evaluators in the social network relationship, the operator can be used to determine the trust value of the indirect trust relationship, so as to determine whether an unknown evaluator can be trusted, so as to obtain a complete social trust relationship between each evaluator.

[0119] Operator P pl The calculation formula is as follows:

[0120]

[0121] in, represents the standardized trust value of the ith edge in the trust propagation chain built based on social network relationships, and σ represents the ith trust value Medium language item σ The subscript value of (q σ ) N,i Trust value Medium language item σ The corresponding probability; g -1 (β) = h βσ It means converting the value β∈[0,1] into a language item in the second language item set H.

[0122] [2] Based on the complete trust relationship in the social network, the expected degree of trust between any two evaluators is calculated using the following formula:

[0123] [3] Calculate the evaluator e s Trust level:

[0124] [4] Calculate the weight of each evaluator:

[0125]

[0126] Among them, ε is a sort, RT ε(l) represents the set {RT1,RT2,…,RT K} the lth largest element in; represents the set {RT1,RT2,…,RT K The sum of the trustworthiness of the first s elements in}, and It represents a fuzzy linguistic quantifier;

[0127] [5] The weighted average (WA) operator is used to aggregate the decision preference matrices of all evaluators to obtain the group decision preference matrix The calculation formula is:

[0128]

[0129] in, Represents the evaluator s The normalized form of the given decision preference matrix.

[0130] In step S4, the consensus of each evaluator is calculated. When the consensus of a certain evaluator does not reach the pre-defined consensus threshold, a personalized feedback mechanism is used to adjust the personal decision preference matrix of the evaluator who has not reached a consensus, and the adjusted group decision preference matrix is ​​obtained.

[0131] [1] A consensus threshold γ is defined in advance to identify the evaluators who have not reached the consensus level among the K evaluators, as well as the corresponding evaluation schemes that have not reached the consensus level and the evaluation elements corresponding to the evaluation schemes. The specific formula is mainly divided into three steps as follows:

[0132] (1) Determine the set of evaluators who have not reached the consensus level: ECH = {s|CI s <γ},

[0133] in, Represents the evaluator s degree of consensus;

[0134] (2) Determine the set of solutions that do not reach the consensus level corresponding to the evaluators who do not reach the consensus level:

[0135] in, Represents the evaluator s Corresponding emergency task planning plan x i degree of consensus;

[0136] (3) Determine the set of elements that do not reach the consensus level, in, Represents the evaluator s Corresponding emergency task planning plan x i For the evaluation index C j degree of consensus.

[0137] [2] Implement a personalized feedback mechanism to determine the evaluators who need to be modified based on the collective APS and the personal preference information of the corresponding positions of the emergency task planning scheme and evaluation indicators.

[0138] In the feedback adjustment process, there is an interactive behavior between the host and the evaluator. In the interactive process, the host, as a leader, determines the adjustment range of the evaluation information, and the evaluator, as a follower, determines the specific personal preference information based on the adjustment range of the evaluation information determined by the host. Then the host reconsiders the adjustment range of the evaluation information based on the personal preference information determined by the evaluator, and the cycle repeats until the Nash equilibrium is reached. Therefore, we use the Stackelberg game model to represent the interactive behavior between the host and the evaluator in the feedback adjustment process.

[0139] From the perspective of the moderator during the interaction, the moderator's goal is to design a compensation strategy that enables all evaluators in the group to reach a pre-defined consensus threshold with minimal adjustment cost:

[0140] Determine personalized feedback parameters for all inconsistent elements based on the collective APS Use this parameter to generate recommended evaluation information The calculation formula is:

[0141]

[0142] in, Represents the evaluator s In emergency mission planning program x i The corresponding evaluation index C j Personalized feedback parameters under

[0143] The adjustment cost considered by the moderator is calculated by measuring the sum of the deviations between the initial preferences of all evaluators and the adjusted preference information, i.e.

[0144]

[0145] in, represents the adjusted preference information,

[0146] From the evaluator's perspective during the interaction, the evaluator's main goal is to adjust the scope based on the evaluation information, i.e. To determine the specific value of the adjusted personal preference information, it is necessary to ensure that each evaluator has the greatest satisfaction during the adjustment process, so as to ensure the public's satisfaction with the evaluation results and evaluation quality. Because the maximum satisfaction among evaluators will maximize the credibility of the evaluation results, the satisfaction calculation formula for each evaluator is:

[0147]

[0148] in, Denote any two evaluators e s and e r The deviation of the adjusted decision matrix, w sr Denote any two evaluators e s and e r The relative weight between

[0149] Therefore, the personalized feedback mechanism that minimizes the adjustment cost of the moderator and maximizes the satisfaction of each evaluator can be summarized as follows:

[0150]

[0151] stmax G HC,

[0152] [3] The WA operator is used to calculate the adjusted group decision preference matrix:

[0153] The calculation formula is:

[0154]

[0155] in, Represents the evaluator s The standardized form of the adjusted decision preference matrix.

[0156] In step S5, according to the adjusted group decision preference matrix, the recommended ranking of the emergency task planning schemes is obtained to determine the optimal emergency task planning scheme; specifically, the following steps are performed:

[0157] Aggregate the evaluation information of all evaluation indicators corresponding to each emergency task planning scheme to obtain the evaluation information of n observation schemes The calculation formula is as follows:

[0158]

[0159] Where ω=(ω1,ω2,…,ω m ) represents the weights corresponding to the pre-given m evaluation indicators;

[0160] Calculate the expected degree corresponding to the evaluation information of each emergency task planning scheme Compare the expected degree of each emergency task planning scheme, and select the observation scheme with the largest expected degree as the optimal emergency task planning scheme; the calculation formula of the expected degree is as follows:

[0161]

[0162] The embodiment of the present invention provides an emergency task planning scheme evaluation system based on a personalized feedback mechanism, including:

[0163] The personal evaluation module is used to obtain the evaluation information of each evaluator corresponding to each evaluation indicator under each emergency task planning scheme according to a number of pre-given emergency task planning schemes, and determine the personal decision preference matrix of each evaluator;

[0164] A trust value acquisition module, used to acquire the social network relationship between each evaluator, and the trust value between each evaluator in the social network relationship;

[0165] A group decision module is used to obtain the weight of each evaluator according to the trust value in the social network relationship; and to obtain the group decision preference matrix by aggregating the individual decision preference matrices of each evaluator through the weights;

[0166] The decision adjustment module is used to calculate the consensus of each evaluator. When the consensus of a certain evaluator does not reach the pre-defined consensus threshold, a personalized feedback mechanism is used to adjust the individual decision preference matrix of the evaluator who has not reached a consensus, and obtain the adjusted group decision preference matrix;

[0167] The scheme recommendation module is used to obtain the recommended ranking of several emergency task planning schemes according to the adjusted group decision preference matrix, and determine the optimal emergency task planning scheme.

[0168] An embodiment of the present invention provides a storage medium storing a computer program for emergency task planning scheme evaluation based on a personalized feedback mechanism, wherein the computer program enables a computer to execute the emergency task planning scheme evaluation method as described above.

[0169] An embodiment of the present invention further provides an electronic device, including:

[0170] one or more processors;

[0171] Memory; and

[0172] One or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, the programs including instructions for executing the emergency mission planning solution evaluation method as described above.

[0173] It can be understood that the emergency task planning scheme evaluation system, storage medium and electronic device based on personalized feedback mechanism provided in the embodiment of the present invention correspond to the emergency task planning scheme evaluation method based on personalized feedback mechanism provided in the embodiment of the present invention. The explanation, examples and beneficial effects of the relevant contents can refer to the corresponding parts in the emergency task planning scheme evaluation method, and will not be repeated here.

[0174] In summary, compared with the prior art, the present invention has the following beneficial effects:

[0175] 1. The advantages of the embodiments of the present invention include at least: on the one hand, it is no longer limited to the premise that the evaluators are independent of each other, and the social network relationship and trust value between each evaluator are taken into account; on the other hand, a personalized feedback mechanism based on the Stackelberg game architecture is proposed, which not only ensures that the group consensus level required for group decision-making is achieved under the condition of minimum adjustment cost and maximum harmony between evaluators, but also takes into account the personal characteristics of the evaluators to achieve group diversity, avoiding the situation of reaching consensus for the sake of reaching consensus.

[0176] 2. If the social network relationship is incomplete, the Archimedean t-norm probability language trust propagation operator P is introduced pl To estimate the indirect trust value between unknown evaluators in the social network relationship, the operator can be used to determine the trust value of the indirect trust relationship, so as to determine whether an unknown evaluator can be trusted, so as to obtain a complete social trust relationship between each evaluator.

[0177] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0178] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An emergency task planning scheme evaluation method based on personalized feedback mechanism, characterized in that: include: According to a number of pre-given emergency task planning schemes, the evaluation information of each evaluator corresponding to each evaluation indicator under each emergency task planning scheme is obtained, and the personal decision preference matrix of each evaluator is determined; Obtaining the social network relationships between the evaluators and the trust values ​​between the evaluators in the social network relationships; Obtaining a weight of each evaluator according to the trust value in the social network relationship; And the individual decision preference matrix of each evaluator is aggregated by weight to obtain the group decision preference matrix; Calculate the consensus of each evaluator. When the consensus of a certain evaluator does not reach the pre-defined consensus threshold, use the personalized feedback mechanism to adjust the individual decision preference matrix of the evaluator who has not reached a consensus, and obtain the adjusted group decision preference matrix; According to the adjusted group decision preference matrix, a recommended ranking of the emergency task planning schemes is obtained to determine the optimal emergency task planning scheme; The step of determining each evaluator's personal decision preference matrix includes: Define n emergency task planning schemes given in advance with X = {x1, x2, …, x n }, each planning scheme corresponds to a predetermined m evaluation index, which is represented by C = {C1, C2, …, C m } indicates that the pre-given first language item set S = {s t |t=0,1,…,τ}, t represents the subscript of the language item, and the total number of language items in the first language item set is τ+1; is assessed by the evaluator s Given an individual decision preference matrix, a set of possible language items is used as the expression form; Represents the evaluator s Given an emergency mission planning solution x i For the evaluation index C j preference information, Representation elements The number of language items in , Represents a language item, In the planning scheme x i For the evaluation index C j Chinese Language The corresponding probability; When there is When the decision preference matrix given by all evaluators is standardized so that the sum of the probabilities corresponding to all language items is equal to 1, the standardized form of preference information is in in, It means that the deviation of the probability and corresponding to 1 and all language items is divided into (1+τ) parts, and Λ represents the preference information The set of all language item subscripts in the first language item set S, Γ = {0, 1, ..., τ}; ρ represents all the items in the first language item set S but not in the preference information The set of all language item subscripts in ; The obtaining of the trust value between each evaluator in the social network relationship includes: T=(t sr ) K×K It represents the trust value between each evaluator in a given social network relationship, and adopts a set of possible degree language items as the expression form; K represents the total number of evaluators in the evaluation process, is a set of possible degree linguistic items, representing the evaluator e s For r Trust value, H = {h σ |σ=0,1,…,χ} is a pre-given set of second language items, σ represents the subscript of the second language item set H, and the total number of language items in the second language item set is χ+1. Represents a language item, Represents the evaluator s For r The language term in the trust value The corresponding probability; When there is When , the trust value between each evaluator is standardized. express, The calculation formula is as follows: in, It means that the deviation of the probability sum corresponding to 1 and all language items is divided into (1+χ) parts evenly, and Λ1 represents the trust value t sr The set of all language item subscripts in , Γ1={0,1,…,χ}; ρ1 represents all items in the second language set H but not in the trust value t sr The set of all language item subscripts in ; If the social network relationship is incomplete, the Archimedean t-norm probability language trust propagation operator P is introduced. pl To estimate the indirect trust value between unknown evaluators in the social network relationship, including: in, i=1,2,…,n represents the standardized trust value of the ith edge in the trust propagation chain built based on social network relationships, and σ represents the ith trust value Medium language item σ The subscript value of (q σ ) N,i Trust value Medium language item σ The corresponding probability; g -1 (β) = h βσ It means converting the value β∈[0,1] into a language item in the second language item set H.

2. The emergency task planning scheme evaluation method according to claim 1, characterized in that: The step of obtaining the weight of each evaluator according to the trust value in the social network relationship includes: According to the trust value in the social network relationship, the expected degree of trust value between any two evaluators is: Calculate the evaluator s Trust level: s=1,2,…,K; Calculate the weight of each evaluator: Among them, ε is a sort, RT ε(l) represents the set {RT1,RT2,…,RT K } the lth largest element in; represents the set {RT1,RT2,…,RT K The sum of the trustworthiness of the first s elements in}, and It represents a fuzzy linguistic quantifier; The individual decision preference matrices of each evaluator are aggregated by weight to obtain a group decision preference matrix, including: The weighted average operator is used to aggregate the decision preference matrices of all evaluators to obtain the group decision preference matrix The calculation formula is: Among them, WA represents the weighted average operator; Represents the evaluator s The normalized form of the given decision preference matrix.

3. The emergency task planning scheme evaluation method according to claim 1, characterized in that: Determining whether the consensus of a certain evaluator does not reach the pre-defined consensus threshold includes: (1) Determine the set of evaluators who have not reached the consensus level: ECH = {s|CI s <γ}, in, Represents the evaluator s The consensus degree of ;γ represents the consensus threshold; (2) Determine the set of solutions that do not reach the consensus level corresponding to the evaluators who do not reach the consensus level: in, Represents the evaluator s Corresponding emergency task planning plan x i degree of consensus; (3) Determine the set of elements that do not reach the consensus level, in, Represents the evaluator s Corresponding emergency task planning plan x i For the evaluation index C j degree of consensus; The personalized feedback mechanism is used to adjust the personal decision preference matrix of the evaluators who have not reached a consensus, specifically based on the Stackelberg game model to represent the interactive behavior between the host and the evaluator in the feedback adjustment process, including: Minimize the host's adjustment costs: Determine personalized feedback parameters for all inconsistent elements based on the collective APS (s,i,j)∈APS, use this parameter to generate recommended evaluation information The calculation formula is: in, Represents the evaluator s In emergency mission planning program x i The corresponding evaluation index C j Personalized feedback parameters under The adjustment cost considered by the moderator is calculated by measuring the sum of the deviations between the initial preferences of all evaluators and the adjusted preference information, i.e. in, represents the adjusted preference information, Maximize the satisfaction of each evaluator: The adjustment range based on the evaluation information, i.e. Determine the specific value of the adjusted personal preference information, and the satisfaction calculation formula for each evaluator is: in, Denote any two evaluators e s and e r The deviation of the adjusted decision matrix, w sr Denote any two evaluators e s and e r The relative weight between s,r=1,2,…,K; The step of obtaining the adjusted group decision preference matrix includes: The WA operator is used to calculate the adjusted group decision preference matrix: The calculation formula is: in, Represents the evaluator s The standardized form of the adjusted decision preference matrix.

4. The emergency task planning scheme evaluation method according to claim 3, characterized in that: Determining the optimal emergency task planning scheme specifically includes: Aggregate the evaluation information of all evaluation indicators corresponding to each emergency task planning scheme to obtain the evaluation information of n observation schemes i=1,2,…,n, The calculation formula is as follows: Where ω=(ω1,ω2,…,ω m ) represents the weights corresponding to the pre-given m evaluation indicators; Calculate the expected degree corresponding to the evaluation information of each emergency task planning scheme i=1,2,…,n, compare the expectation of each emergency task planning scheme, and select the observation scheme with the largest expectation as the optimal emergency task planning scheme; the calculation formula of expectation is as follows:

5. An emergency task planning scheme evaluation system based on personalized feedback mechanism, characterized in that: The method for evaluating an emergency task planning scheme according to claim 1 comprises: The personal evaluation module is used to obtain the evaluation information of each evaluator corresponding to each evaluation indicator under each emergency task planning scheme according to a number of pre-given emergency task planning schemes, and determine the personal decision preference matrix of each evaluator; A trust value acquisition module, used to acquire the social network relationship between each evaluator, and the trust value between each evaluator in the social network relationship; A group decision module is used to obtain the weight of each evaluator according to the trust value in the social network relationship; and to obtain the group decision preference matrix by aggregating the individual decision preference matrices of each evaluator through the weights; The decision adjustment module is used to calculate the consensus of each evaluator. When the consensus of a certain evaluator does not reach the pre-defined consensus threshold, a personalized feedback mechanism is used to adjust the individual decision preference matrix of the evaluator who has not reached a consensus, and obtain the adjusted group decision preference matrix; The scheme recommendation module is used to obtain the recommended ranking of several emergency task planning schemes according to the adjusted group decision preference matrix, and determine the optimal emergency task planning scheme.

6. A storage medium, characterized in that: It stores a computer program for emergency task planning scheme evaluation based on a personalized feedback mechanism, wherein the computer program enables a computer to execute the emergency task planning scheme evaluation method as described in any one of claims 1 to 4.

7. An electronic device, characterized in that: include: one or more processors; Memory; as well as One or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the programs include a program for executing the emergency mission planning scheme evaluation method according to any one of claims 1 to 4.

Images