Data integrity auditing and matching method based on psychological perception and preference

Abstract

The invention requests to protect a data integrity audit matching method based on psychological perception and preference, and the method comprises the steps: S1, calculating subjective and objective weights of each evaluation index, calculating subjective and objective combined weights of the indexes based on the calculated subjective and objective combined weights, and then calculating the weight of each audit scheme by using an improved Topsis comprehensive evaluation method; s2, according to the user preference set, calculating user preference weights by using a triangular fuzzy number and a Topsis method, and gathering the weights of the two; s3, according to the demand preference of the auditing task and the execution preference of the auditing resource, the comprehensive preference values of the two parties are quantitatively calculated, and a preference utility value matrix of the two parties is calculated; s4, based on an improved disappropriation theory, calculating a perception utility value matrix of the two parties; and S5, carrying out standardization processing on the perception utility value matrix, constructing a double-target optimization model by taking matching of standard perception utility values of the two parties as an optimization target, and carrying out solving. According to the method, under the condition of considering psychological factors and preferences of two matching parties, the comprehensive satisfaction degree of the two matching parties is effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of data integrity audit and bilateral matching, and in particular to a bilateral matching method of user audit tasks and audit schemes (resources) before data integrity audit in consideration of the psychological perception and preference of both parties. Background Art

[0002] In recent years, the field of cloud computing has made significant progress, providing flexible and reliable computing and storage solutions for many users. With the rapid development of 5G and mobile network technology, the diversity and complexity of online applications have been greatly improved, including interactive online games, virtual reality (VR), video analysis, and natural language processing. However, these online applications are very sensitive to latency, and there is often high and unstable latency between cloud services and end users. Therefore, application service providers are in urgent need of deploying applications near users to meet the increasingly stringent requirements for latency. Edge computing is considered to be a key technology for reducing latency. It is a distributed computing model whose core concept is to push computing power and data storage to the edge of the network to reduce latency during data transmission and improve system response speed. In the mobile edge computing (MEC) scenario, in order to enable terminal devices to directly access data, edge nodes usually pre-load part of the data or services from the central cloud server. However, in the edge computing environment, the edge server may be subject to internal attacks or external failures, which may cause the integrity of edge data to be damaged, so the security of edge data is also receiving increasing attention.

[0003] Data integrity detection technology is an effective means to check the integrity of data. Users can use this technology to verify the integrity of data stored at the edge to ensure data security and availability. However, existing data integrity auditing solutions often focus on improving audit efficiency, the reliability of audit results, and subsequent functional expansion, while ignoring the adaptability of a single audit strategy and not considering how to match the most appropriate auditing solution to the needs and preferences of each task when processing multiple user audit tasks. On the one hand, a single data integrity auditing solution is difficult to cope with the rapid changes in massive data in edge environments and the diversity of user preferences. On the other hand, in current bilateral matching research, the psychological impact of individuals on matching results and the preference needs of both parties are often easily overlooked. Therefore, how to further improve the comprehensive satisfaction of both parties while considering the psychological perception and preferences of both parties is an urgent problem to be solved.

[0004] Therefore, the present invention proposes a data integrity audit matching method based on psychological perception and preference.

[0005] After searching, the application publication number CN117114280A belongs to the field of bilateral matching. The method includes proposing a bilateral matching method of R&D institution team and scientific and technological talents based on prospect theory, and the steps are as follows: constructing a first-stage elimination matching method for new R&D institution teams and scientific and technological talents, and constructing a first-stage elimination matching model M-1; constructing a second-stage selection matching for new R&D institution teams and scientific and technological talents, and constructing a second-stage selection matching model M-2. The invention considers the reference point and the psychological factors of decision makers, combines prospect theory with bilateral matching, and considers the two stages of bilateral matching. The first stage considers benefits and costs, and the second stage considers the satisfaction of both the team and the talent.

[0006] The difference from the above method is that the present invention, while achieving a satisfactory bilateral match between audit tasks and audit resources, describes the attributes of audit tasks and resource solutions, and quantitatively calculates the comprehensive preference values ​​of both parties according to the demand preferences of audit tasks and the execution preferences of audit resources, and considers the preference needs of both parties in the matching process. At the same time, the disappointment theory adopted is improved. In constructing the pleasure function, this patent considers the disappointment avoidance effect, that is, introduces the disappointment avoidance parameter to correct the pleasure value, and considers the subjective characteristics of both parties in the matching when calculating the perceived utility value matrix, rather than completely rational individuals. Finally, the optimization goal is to maximize the standard perceived utility value of both parties, and the dual-objective optimization model is converted into a single-objective optimization model by using a linear weighted method for solution. Summary of the invention

[0007] The present invention aims to solve the above problems of the prior art. A data integrity audit matching method based on psychological perception and preference is proposed. The technical solution of the present invention is as follows:

[0008] A data integrity audit matching method based on psychological perception and preference, comprising the following steps:

[0009] S1. Calculate the subjective and objective weights ω of each evaluation index s and ω o , and calculate the combined weight ω of each indicator based on the subjective and objective weights of the indicators c , and then use the improved Topsis comprehensive evaluation method to calculate the weight of each audit plan ω a ;

[0010] S2. Construct a weighted judgment matrix H based on the user preference sets y[M] and y[N][M]. ij According to the weighted judgment matrix, the comprehensive non-fuzzy evaluation matrix E is calculated using the mean area method. ij , using Topsis to calculate the weights of each audit scheme considering user preference requirements b , Gather ω aand ω b The weights of the two are used to obtain the final weight of each audit scheme ω f ;

[0011] S3. According to the final weight of each audit scheme calculated, based on the preference requirements of both parties, the comprehensive preference value tp of audit tasks to audit resources and audit resources to audit tasks is quantitatively calculated. ij and rp ij , and construct the preference order matrix TPO and RPO, and then convert the preference order matrix into the preference utility value matrix TPU and RPU according to the preference utility function;

[0012] S4. Using the improved disappointment theory (i.e., the preference order values ​​of both parties are quantified and calculated, and at the same time, when constructing the pleasure function, this patent takes into account the disappointment avoidance effect, and when calculating the perceived utility value matrix of both parties, the subjective characteristics of both parties are considered to match, rather than completely rational individuals), respectively, the disappointment function and the pleasure function form considering the disappointment avoidance effect are constructed, and the comprehensive loss value matrix d(r ij ), d`(t ij ) and the comprehensive happiness value matrix e(r ij ), e`(t ij ), and then modify the preference utility value to obtain the perceived utility value matrix u(r ij ),u`(t ij );

[0013] S5. Standardize the perceived utility value matrix of both parties to obtain the standard perceived utility value. and Maximization is taken as the optimization goal, a dual-objective optimization model is constructed, and then solved.

[0014] Furthermore, in step S1, the subjective weight and objective weight of the evaluation index and the subjective and objective combined weight are calculated, and then the weight of each audit scheme is calculated based on the improved Topsis comprehensive evaluation method, specifically including:

[0015] S11. Based on the data results of the twelve expert questionnaires obtained from the survey, the improved G1 ordinal relationship analysis method was used to calculate the subjective weights of the evaluation indicators. That is, the opinions of multiple experts were comprehensively considered when using the G1 ordinal relationship analysis method to calculate the subjective weights of the indicators. This is more in line with the objectivity of group decision-making and avoids the deviation that may be caused by the subjective empowerment of a single expert.

[0016] S12. According to the attribute set of the data integrity audit scheme, the objective weight of the evaluation index is calculated using the improved Critic objective weight method. That is, when using the Critic method to calculate the indicator variability, the influence of the average value on the indicator variability is considered, making the calculation of the coefficient of variation more reasonable and accurate.

[0017] S13. Use linear weighting method to calculate the subjective and objective combined weight of each indicator.

[0018] S14. According to the subjective and objective combined weights, the improved Topsis comprehensive evaluation method is used to calculate the weight of each audit plan. That is, when calculating the weighted Euclidean distance between each audit plan and the positive ideal value and the negative ideal value, the subjective and objective combined weights of the indicators are combined, which effectively improves the rationality and accuracy of the weight calculation results.

[0019] Furthermore, in step S2, the weights of each audit scheme considering the user's preference requirements are calculated and the final weights of each audit scheme are obtained by combining the weights of the two, which specifically includes:

[0020] S21. Based on triangular fuzzy numbers, the user language evaluation is fuzzified and quantified, and a weighted judgment matrix is ​​calculated according to the user preference set to obtain a comprehensive fuzzy evaluation matrix;

[0021] S22. Use the mean area method to defuzzify the comprehensive fuzzy evaluation matrix to obtain a comprehensive non-fuzzy evaluation matrix, and use the Topsis comprehensive evaluation method to calculate the weights of each audit scheme considering user preference requirements;

[0022] S23. Using the linear weighting method, the weights of the two are combined to calculate the final weight of each audit plan.

[0023] Furthermore, in step S3, the comprehensive preference values ​​of the audit task to the audit resource and the audit resource to the audit task are quantified and the preference utility values ​​of both parties are calculated based on the preference utility function, which specifically includes:

[0024] S31. Describe the attributes of the user audit task and audit resource, quantify and calculate the preference values ​​of each party according to the preference requirements of both parties, and use the linear weighted method to calculate the comprehensive preference value of both parties, and calculate the preference sequence value according to the comprehensive preference value of the individual;

[0025] S32. According to the preference order matrix of the matching parties, a preference utility function is constructed, and the preference utility value is calculated to obtain the preference utility value matrix of the audit task to the audit resource and the preference utility value matrix of the audit resource to the audit task.

[0026] Furthermore, in step S4, the comprehensive loss value and comprehensive pleasure value matrix of the matching parties are calculated and the preference utility value is corrected to obtain a perceived utility value matrix, which specifically includes:

[0027] S41. Based on the improved disappointment theory, construct the disappointment function and calculate the subject A i For subject B j Matches instead of B k The loss value of the audit task to the audit resource is d(r ij ) is an arithmetic mean. Similarly, construct a happiness function and calculate the comprehensive happiness value matrix of the matching parties under the consideration of the disappointment avoidance effect.

[0028] S42. According to the comprehensive pleasure value and comprehensive loss value matrix of the matching parties, the preference utility value matrix of both parties is modified to obtain the perceived utility value matrix of the audit task to the audit resources and the perceived utility value matrix of the audit resources to the audit task.

[0029] Furthermore, in step S5, the perceived utility value matrix is ​​standardized, and a dual-objective optimization model is constructed and solved with maximizing the standard perceived utility value as the optimization goal, which specifically includes:

[0030] S51. Standardize the perceived utility value, take maximizing the standard perceived utility value of both parties as the optimization goal, build a dual-objective optimization model, and use a linear weighted method to convert the dual-objective optimization model into a single-objective optimization model;

[0031] S52. Use KM algorithm to find the optimal matching result.

[0032] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the user data integrity audit matching method based on psychological perception and preference as described in any one of the items is implemented.

[0033] A non-transitory computer-readable storage medium stores a computer program thereon, wherein when the computer program is executed by a processor, the user data integrity audit matching method based on psychological perception and preference as described in any one of the items is implemented.

[0034] A computer program product comprises a computer program, wherein when the computer program is executed by a processor, the method for user data integrity audit matching based on psychological perception and preference as described in any one of the items is implemented.

[0035] The advantages and beneficial effects of the present invention are as follows:

[0036] 1. In step S1 of the present invention, when the subjective weights of the evaluation indicators are calculated using the improved G1 ordinal relationship analysis method, the opinions of multiple experts are comprehensively considered, which is more in line with the objectivity of group decision-making and avoids the deviation that may be caused by the subjective weighting of a single expert; when the Critic method is used to calculate the indicator variability, the influence of the average value on the indicator variability is considered, so that the calculation of the coefficient of variation is more reasonable and accurate; finally, the weights of each audit plan are calculated by combining the calculated subjective and objective combined weights, and the improved Topsis comprehensive evaluation method is used to effectively improve the rationality and accuracy of the weight calculation results.

[0037] 2. In step S3 of the present invention, when calculating the final weight of the audit scheme, the user's preference weight is further considered, that is, the weights of each audit scheme considering the user's preference requirements calculated in step S2 according to the user preference set using triangular fuzzy numbers and Topsis are gathered in the calculation of the final weight.

[0038] 3. In step S4 of the present invention, by describing the attributes of the audit task and resource solution, the comprehensive preference value of both parties is quantitatively calculated according to the audit task demand preference and the audit resource execution preference, and the preference requirements of both parties are considered in the bilateral matching process. In addition, in the subsequent matching, the psychological factors of both parties are considered by using the improved disappointment theory.

[0039] 4. In step S6 of the present invention, the multi-objective optimization problem of bilateral matching is cleverly converted into a single-objective optimization problem by means of linear weighting, and a dual-objective optimization model is constructed to find the optimal match. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a flow chart of the overall preferred embodiment provided by the present invention;

[0041] Figure 2 This is the system architecture diagram;

[0042] Figure 3 It is a flow chart of the audit scheme evaluation model considering user preferences based on triangular fuzzy numbers;

[0043] Figure 4 It is a flow chart of the bilateral matching strategy between audit tasks and audit resources based on psychological perception and preference. DETAILED DESCRIPTION

[0044] The following will describe the technical solutions in the embodiments of the present invention in detail in conjunction with the accompanying drawings in the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention.

[0045] The technical solution of the present invention to solve the above technical problems is:

[0046] 1. In view of the fact that the adaptability of a single audit strategy is ignored in the existing data integrity audit strategy, and that a single audit strategy is difficult to adapt to the rapid changes of massive data in edge environments and the problem of data diversity, this paper proposes an audit scheme evaluation model based on triangular fuzzy numbers that considers user preferences, calculates the weight of each audit scheme, and ranks them according to the weights to screen out audit schemes that are more in line with user preferences.

[0047] 2. The existing data integrity audit strategy does not consider how to match the most appropriate audit scheme according to the needs and preferences of each task when dealing with multiple user audit tasks before the audit. In addition, in the current bilateral matching algorithm, the psychological impact of the matching results on individuals and the preference needs of both parties are often overlooked. Therefore, this paper proposes a bilateral matching strategy for audit tasks and audit resources based on psychological perception and preference. Using the improved disappointment theory, the standard perceived utility value matrix of audit tasks to audit resources and audit resources to audit tasks is calculated respectively. Finally, the dual-objective optimization model is constructed with the maximization of the standard perceived utility value (satisfaction) of both parties as the optimization goal, and the solution is obtained.

[0048] like Figure 1 As shown, the present invention provides a data integrity audit matching method based on psychological perception and preference, characterized by comprising the following steps:

[0049] S1. Calculate the subjective and objective weights ω of each evaluation index s and ω o , and calculate the combined weight ω of each indicator based on the subjective and objective weights of the indicators c , and then use the improved Topsis comprehensive evaluation method to calculate the weight of each audit plan ω a ;

[0050] S2. Construct a weighted judgment matrix H based on the user preference sets y[M] and y[N][M] ij According to the weighted judgment matrix, the comprehensive non-fuzzy evaluation matrix E is calculated using the mean area method. ij , using Topsis to calculate the weights of each audit scheme considering user preference requirements b , Gather ω a and ω b The weights of the two are used to obtain the final weight of each audit scheme ω f .

[0051] S3. According to the final weight of each audit scheme calculated, based on the preference requirements of both parties, the comprehensive preference value tp of audit tasks to audit resources and audit resources to audit tasks is quantitatively calculated. ij and rp ij, and construct the preference order matrices TPO and RPO, and then according to the preference utility function, convert the preference order matrices into preference utility value (satisfaction) matrices TPU and RPU.

[0052] S4: Using the improved disappointment theory, the disappointment function and the pleasure function considering the disappointment avoidance effect are constructed respectively, and the comprehensive loss value matrix d(r ij ), d`(t ij ) and the comprehensive happiness value matrix e(r ij ), e`(t ij ), and in consideration of the psychological perception of both parties, the preference utility value is modified to obtain the perceived utility value matrix u(r ij ),u`(t ij ).

[0053] S5: Standardize the perceived utility value matrix of both parties to obtain the standard perceived utility value. and Maximization is taken as the optimization goal, a dual-objective optimization model is constructed, and then solved.

[0054] In this example, in step S1, the subjective weight ω of each evaluation index is calculated using the following method: s :

[0055] S11: First, we construct an evaluation index system for the comprehensive evaluation model. This paper adopts the following seven indicators as evaluation indicators, namely, computational efficiency, storage efficiency, time complexity, communication efficiency, security, accuracy and functionality. In order to avoid the possible deviation caused by the subjective weighting of a single expert, this paper uses the G1 order relationship analysis method to calculate the subjective weight ω s The opinions of many experts were considered. The expert scoring method was used to assign 90 points, 80 points, 70 points, 60 points, and 50 points to each indicator according to the categories of very important, important, generally important, unimportant, and not considered. Then, based on the data from the twelve expert questionnaires, the importance of each evaluation indicator was calculated in the following way. For example, the score calculation process of the "computational efficiency index" was: 9 / 12×90+3 / 12×80+0 / 12×70+0 / 12×60+0 / 12×50=87.5 points. Then, according to the ranking of the indicator scores, the order relationship between the indicators was determined, for example After determining the order relationship between the indicators, calculate two adjacent indicators C j and C j+1 The relative importance of β , r β The larger the value, the higher the index C. j Ratio Index C j+1The greater the importance, the more important it is. Since this paper takes into account the subjective will of multiple experts, the relative importance of the indicators assigned by experts k = 1, 2, ..., n is expressed as This paper selected 6 experts from the above 12 experts to carry out this work, and the relative importance of the final indicator was r β The calculation formula is as follows:

[0056]

[0057] According to the decision maker's β The subjective weight of the indicator is calculated by the following two formulas:

[0058]

[0059] ω β-1 =ω β *r β #(3)

[0060] where β = n, n-1, …, 2, and

[0061] In this example, in step S1, the objective weight ω of each evaluation index is calculated using the following method: o :

[0062] S12: Considering that the evaluation indicators adopted in this paper are all qualitative indicators, the objective weight ω of the indicator is calculated using the Critic method. o Before that, it is necessary to quantify the qualitative indicators. This paper adopts the 1-9 scale method, and assigns each indicator with 9 points, 7 points, 5 points, 3 points, and 1 point according to excellent, good, medium, poor, and very poor in the form of expert scoring. 12 experts are selected to score, and the qualitative indicators of each plan are quantified. For example, the final score of audit plan 1 in terms of computational efficiency is: (5+9+7+3+3+1+1+9+7+3+5+9) / 12=5.17 points. According to the data from the twelve expert questionnaires, the original data matrix X of i samples to be evaluated and j evaluation indicators is calculated. In order to eliminate the influence of different dimensions on the evaluation results, this paper uses the following formula to perform positive processing of indicators:

[0063]

[0064] Calculate the indicator variability. This paper takes into account the influence of the indicator average value on the indicator variability value. In the process of calculating the indicator variability, the coefficient of variation is introduced Instead of the indicator variability S j The larger the coefficient of variation, the greater the numerical difference of the indicator, the stronger the evaluation strength of the indicator itself, and the higher the weight. The formula for calculating the coefficient of variation is as follows:

[0065]

[0066] in, Indicates the average value of the indicator.

[0067] Then calculate the index conflict, index conflict R j The calculation formula is as follows:

[0068]

[0069] Among them, r ij It represents the correlation coefficient between the i-th indicator and the j-th indicator. The stronger the correlation coefficient with other indicators, the less conflict this indicator has with other indicators and the lower its weight.

[0070] Furthermore, the indicator information C is calculated based on the indicator variability and indicator conflict. j , according to the amount of information C j Calculate the objective weight of the indicator ω o , the calculation formula is as follows:

[0071] C j =S j *R j #(9)

[0072]

[0073] Among them, the indicator information volume C j The larger it is, the greater the role of the jth evaluation indicator in the entire evaluation method and the higher its weight.

[0074] In this example, in step S1, the subjective and objective combined weight ω of each evaluation index is calculated using the following method: c :

[0075] S13: This paper uses a linear weighted method to calculate the subjective and objective combined weights of the evaluation indicators. The calculation formula is as follows:

[0076]

[0077] In this example, in step S1, the weight ω of each audit scheme is calculated using the following method: a :

[0078] S14: According to the subjective and objective combined weights, the weight of each audit plan is calculated using the improved Topsis comprehensive evaluation method. First, the optimal ideal value and the worst ideal value of each indicator of the audit plan are determined. The calculation formula is as follows:

[0079]

[0080] Then calculate the weighted Euclidean distance between each solution and the optimal ideal value and the worst ideal value. The calculation formula is as follows:

[0081]

[0082] According to the weighted Euclidean distance, the scores of each solution when it is not normalized are calculated. The calculation formula is as follows:

[0083]

[0084] After normalization, the weights of each audit scheme are obtained: a , the calculation formula is as follows:

[0085]

[0086] In this example, in step S2, the weights ω of each audit scheme considering user preference requirements are calculated using the following method: b :

[0087] S21: According to the user preference set, the user's language evaluation is fuzzified and quantified using triangular fuzzy numbers. The relationship between the user's language evaluation level and the triangular fuzzy number used in this paper is shown in the following table:

[0088] excellent good medium Difference Very bad (0.8,0.9,1) (0.6,0.7,0.8) (0.4,0.5,0.6) (0.2,0.3,0.4) (0.1,0.2,0.3)

[0089] According to the above table, the fuzzy language evaluation of the indicators under each audit scheme by the user is converted into a triangular fuzzy number, which is recorded as: R ij =(a ij ,b ij ,c ij ), thus, all fuzzy evaluation values ​​of the user for N audit schemes can be expressed by the triangular fuzzy matrix R = [R ij ] is represented. Then the triangular fuzzy evaluation matrix is ​​normalized to obtain the normalized matrix The calculation formula is as follows:

[0090]

[0091] Secondly, the user's evaluation of the importance of each indicator is converted into a triangular fuzzy number, and the triangular fuzzy weight matrix W = [ω j ], normalize it and get the normalized matrix The calculation formula is as follows:

[0092]

[0093] According to the user evaluation triangular fuzzy matrix and triangular fuzzy weight matrix, a weighted judgment matrix H = (H ij ),in The calculation formula is as follows:

[0094]

[0095] S22: Then the weighted judgment matrix is ​​defuzzified using the mean area method to obtain a comprehensive non-fuzzy evaluation matrix E = (e ij ), the calculation formula is as follows:

[0096]

[0097] S23: Therefore, according to the comprehensive non-fuzzy evaluation matrix, the Topsis comprehensive evaluation method is used to calculate the weights of each audit scheme considering user preference requirements b Finally, the final weight ω of each audit scheme is calculated using a linear weighted method f :

[0098]

[0099] In this example, in step S3, the following method is used to quantify the comprehensive preference values ​​of audit tasks to audit resources and audit resources to audit tasks, and to calculate the preference utility values ​​of both parties based on the preference utility function:

[0100] S31. Describe the attributes of task requirements and resource solutions, and user audit task m j (j=1,2,…,m), using a two-dimensional vector T(t w ,t cos ) indicates that the audit strategy (resource) n i (i=1,2,…,n), using the two-dimensional vector T(r w ,r cos ), then the comprehensive preference value of both parties is quantified based on the audit task demand preference and the audit resource execution preference. The calculation formula is as follows:

[0101] The audit task's preference for audit resources in terms of weight is that the weight of the required resources must meet the requirements of their own weight. The higher the resource weight, the more preferred it is by the audit task:

[0102]

[0103] The audit task's preference for audit resources in terms of service price is that the price of the required resources is within the range that the audit task can afford. The lower the price of the resources, the more preferred the task is:

[0104]

[0105] The audit resource's preference for the weight of the audit task is based on the weight required by the task. The lower the weight required by the task, the better, indicating that the more tasks the resource can meet, the more preferred it is:

[0106]

[0107] The audit resource's preference for the service price of audit tasks is based on the price required by the task. The lower the price required by the task, the better, indicating that the resource can meet more tasks and is more preferred by the resource:

[0108]

[0109] Then, the linear weighted method is used to calculate the comprehensive preference values ​​of both parties, and the preference sequence is calculated based on the comprehensive preference values ​​of individuals. The calculation method of the comprehensive preference value of audit tasks for audit resources is:

[0110]

[0111] Similarly, the calculation method of the comprehensive preference value of audit resources for audit tasks is:

[0112]

[0113] S32: According to the comprehensive preference values ​​of both parties, the preference order matrices TPO and RPO of both parties are obtained. In order to better represent the preference degree of a subject for another subject, and decrease with the increase of the preference order, this paper constructs the following preference utility function, converts the preference order value into the preference utility value, and obtains the preference utility value matrices TPU and RPU of both parties:

[0114]

[0115] In this example, in step S4, the following method is used to calculate the comprehensive loss value and comprehensive pleasure value matrix of the matching parties and modify the preference utility value to obtain the perceived utility value matrix u(r ij ),u`(t ij ):

[0116] S41: In order to consider the psychological factors of both parties in the matching process, this paper constructs the following loss function based on the improved disappointment theory:

[0117] D(x)=1-α x #(30)

[0118] Among them, α represents the degree of concavity of the disappointment function. The larger α is, the smaller the subject's disappointment is, and the smaller α is, the greater the subject's disappointment is. Then calculate subject A i For subject B i Matches instead of Bk Matching loss value d k (r ij ), the calculation formula is as follows:

[0119]

[0120] Then the comprehensive loss value of audit resources to audit tasks is D(r ij ) is calculated as follows:

[0121]

[0122] When constructing the pleasure function, this paper takes into account the loss aversion effect and introduces the loss aversion parameter γ to correct the pleasure value. The form of the pleasure function is as follows:

[0123] E(X) = γ(1-β x )#(33)

[0124] Among them, β represents the degree of concavity of the pleasure function. The larger β is, the smaller the subject's pleasure is, and the smaller β is, the greater the pleasure is. Then calculate subject A i For subject B i Matches instead of B k Matching happiness value e k (r ij ), the calculation formula is as follows:

[0125]

[0126] Then the comprehensive pleasantness value of audit task to audit resources is e(r ij ) is calculated as follows:

[0127]

[0128] S42: Calculate the audit resource preference utility value, comprehensive loss value and comprehensive pleasure value of subject A based on the audit task. i For subject B j The matching perceived utility value u(r ij ), that is, the modified preference utility value under the premise of considering the psychological perception of both parties. On this basis, this paper introduces the parameter θ, which takes into account the subjective will of both parties, rather than a completely rational subject, and is more in line with objective reality. The calculation formula is as follows:

[0129] u(r ij )=v(r uj )-θ*d(r ij )+(1-θ)*e(r ij )#(36)

[0130] Similarly, the perceived utility value of audit resources to audit tasks is calculated as:

[0131] u`(t ij )=v(t ij )-θ*d`(t ij )+(1-θ)*e`(t ij )#(37)

[0132] In this example, in step S5, the following method is used to standardize the perceived utility value matrix, and the dual-objective optimization model is constructed and solved with the maximization of the standard perceived utility values ​​of both parties as the optimization goal:

[0133] S51: This paper uses the following formula to standardize the perceived utility values ​​of both parties, and uses a linear weighted approach to convert the dual-objective optimization model into a single-objective optimization model. The calculation formula is as follows:

[0134]

[0135] Among them, x ij represents a 0-1 variable, and Represent the standard perceived utility value (satisfaction) matrix of audit tasks and audit resources respectively. Fall in love with r ij is the column maximum or in the matrix Falling in love ij When it is the maximum value of the row, the corresponding x ij Takes 1, otherwise 0. ij represents the coefficient matrix calculated based on the standard perceived utility value matrix of both parties, and ω1+ω2=1.

[0136] S52: This paper adopts the KM algorithm to find the optimal matching result. The optimal result is: max Z, that is, the comprehensive satisfaction of both parties reaches the highest.

[0137] The following is a specific example to further illustrate:

[0138] According to the data results of the expert questionnaire, the importance score of each evaluation indicator is calculated, and finally the ranking is based on the score. The order relationship between the indicators is as follows:

[0139] Computational efficiency (CE) > Storage efficiency (SE) > Time complexity (TC) > Communication efficiency (COE) > Security (SEC) > Accuracy (AC) > Functionality (FN)

[0140] The relative importance of the final indicator r β The calculation is as follows:

[0141] <![CDATA[r1]]> <![CDATA[r2]]> <![CDATA[r3]]> <![CDATA[r4]]> <![CDATA[r5]]> <![CDATA[r6]]> 1.50 1.40 1.33 1.23 1.20 1.16

[0142] Similarly, based on the data results of the expert questionnaire, the subjective weights of each indicator ω are calculated. s , objective weight ω o , combined weight ω c As shown below:

[0143] CE SE TC COE SEC AC FN <![CDATA[ω s ]]> 0.308 0.206 0.147 0.110 0.090 0.075 0.064 <![CDATA[ω o ]]> 0.204 0.098 0.175 0.108 0.180 0.115 0.120 <![CDATA[ω c ]]> 0.256 0.152 0.161 0.109 0.135 0.095 0.092

[0144] Then, according to the user preference set, the weights of various audit schemes are calculated a ,ω b ,ω f As shown below:

[0145] Solution 1 Solution 2 Option 3 Option 4 <![CDATA[ω a ]]> 0.063 0.143 0.136 0.141 <![CDATA[ω b ]]> 0.066 0.150 0.142 0.150 <![CDATA[ω f ]]> 0.064 0.146 0.139 0.145 Option 5 Option 6 Option 7 Option 8 <![CDATA[ω a ]]> 0.141 0.127 0.119 0.130 <![CDATA[ω b ]]> 0.133 0.126 0.113 0.120 <![CDATA[ω f ]]> 0.137 0.126 0.116 0.125

[0146] Assuming that in a matching pair of audit tasks and audit resources, the number of audit tasks is 11 and the number of audit resources is 8, according to the preference requirements of both parties, after quantitatively calculating the comprehensive preference values ​​of both parties, the preference order of audit tasks to audit resources is obtained as shown in the following table:

[0147] <![CDATA[t1]]> <![CDATA[t2]]> <![CDATA[t3]]> <![CDATA[t4]]> <![CDATA[t5]]> <![CDATA[t6]]> <![CDATA[t7]]> <![CDATA[t8]]> <![CDATA[t9]]> <![CDATA[t 10 ]]> <![CDATA[t 11 ]]> <![CDATA[r1]]> 1 1 3 1 1 1 3 1 1 1 1 <![CDATA[r2]]> 6 3 6 2 2 3 6 2 2 6 2 <![CDATA[r3]]> 5 4 5 3 3 4 5 3 3 2 3 <![CDATA[r4]]> 8 5 8 4 8 5 8 4 4 8 4 <![CDATA[r5]]> 2 6 1 5 4 6 1 5 5 3 5 <![CDATA[r6]]> 3 7 2 6 5 7 2 6 6 4 6 <![CDATA[r7]]> 7 2 7 7 6 2 7 7 7 7 7 <![CDATA[r8]]> 4 8 4 8 7 8 4 8 8 8 8

[0148] The results of the preference order of audit resources for audit tasks are shown in the following table:

[0149] <![CDATA[t1]]> <![CDATA[t2]]> <![CDATA[t3]]> <![CDATA[t4]]> <![CDATA[t5]]> <![CDATA[t6]]> <![CDATA[t7]]> <![CDATA[t8]]> <![CDATA[t9]]> <![CDATA[t 10 ]]> <![CDATA[t 11 ]]> <![CDATA[r1]]> 9 1 10 2 11 8 3 4 5 6 7 <![CDATA[r2]]> 9 4 10 1 11 8 7 5 2 3 6 <![CDATA[r3]]> 9 4 10 1 11 7 8 6 2 3 5 <![CDATA[r4]]> 9 4 7 1 11 10 6 5 2 3 8 <![CDATA[r5]]> 9 6 10 1 11 4 8 7 2 5 3 <![CDATA[r6]]> 9 6 10 1 11 5 8 7 2 3 4 <![CDATA[r7]]> 9 4 10 1 11 8 6 5 3 2 7 <![CDATA[r8]]> 9 4 10 1 11 7 8 6 2 3 5

[0150] According to the preference order matrix, the standard perceived utility value matrix of audit tasks to audit resources and the standard perceived utility value matrix of audit resources to audit tasks are calculated based on the improved disappointment theory. Based on the above two matrices, the coefficient matrix f ij as follows:

[0151] <![CDATA[t1]]> <![CDATA[t2]]> <![CDATA[t3]]> <![CDATA[t4]]> <![CDATA[t5]]> <![CDATA[t6]]> <![CDATA[t7]]> <![CDATA[t8]]> <![CDATA[t9]]> <![CDATA[t 10 ]]> <![CDATA[t 11 ]]> <![CDATA[r1]]> 0.993 1.0 0.5 0.5 0.5 0.609 1.0 0.623 0.958 1.0 1.0 <![CDATA[r2]]> 0.331 0.633 0.176 0.208 0.240 0.278 0.666 0.716 0.726 0.364 0.463 <![CDATA[r3]]> 0.241 0.590 0.195 0.116 0.145 0.289 0.585 0.621 0.617 0.572 0.485 <![CDATA[r4]]> 0.134 0.429 0.211 0.568 0.083 0.243 0.551 0.573 0.477 0.423 0.352 <![CDATA[r5]]> 0.057 0.035 0.341 0.037 0.536 0.696 0.384 0.3 0.494 0.430 0.744 <![CDATA[r6]]> 0.538 0.021 0.254 0.022 0.238 0.402 0.504 0.164 0.459 0.376 0.634 <![CDATA[r7]]> 0.521 0.361 0.515 0.005 0.031 0.506 0.496 0.116 0.515 0.469 0.454 <![CDATA[r8]]> 0.510 0.210 0.210 0 0.025 0.567 0.476 0.125 0.511 0.397 0.517

[0152] Based on the above coefficient matrix, in order to maximize the value of the objective function, the optimal matching results are: (r6, t1), (r7, t3), (r4, t4), (r8, t6), (r3, t8), (r2, t9), (r1, t 10 ),(r5,t 11 ), there is no resource match for tasks t2, t5, and t7, and the overall satisfaction of both parties with the matching results is: 5.28.

[0153] Aiming at the problem of the adaptability of a single audit strategy being ignored in the existing data integrity audit strategy, the present invention proposes an audit scheme evaluation model based on triangular fuzzy numbers that considers user preferences. This paper uses the improved G1 order relationship analysis method to calculate the subjective weights of the evaluation indicators, and comprehensively considers the opinions of multiple experts, which is more in line with the objectivity of group decision-making and avoids the bias that may be caused by the subjective weighting of a single expert; when using the Critic method to calculate the indicator variability, the influence of the average value on the indicator variability is considered, making the calculation of the coefficient of variation more reasonable and accurate; finally, combined with the calculated subjective and objective combined weights, the improved Topsis comprehensive evaluation method is used to calculate the weights of each audit scheme, which effectively improves the rationality and accuracy of the weight calculation results. In addition, compared with other evaluation models, the present invention takes into account the personalized demand weights of users in the comprehensive evaluation, which can more accurately reflect the actual needs and preferences of users, provide more personalized services, and further improve the quality of decision-making. At the same time, the existing data integrity audit strategy does not consider how to match the most appropriate audit scheme according to the needs and preferences of each task when dealing with multiple user audit tasks before the audit. In addition, in the current bilateral matching algorithm, the psychological impact of the matching results on individuals and the preference needs of both parties are often overlooked. This paper proposes a bilateral matching method based on psychological perception and preference. Based on the improved disappointment theory, the standard perceived utility value matrix of audit tasks to audit resources and audit resources to audit tasks is calculated respectively. Finally, the dual-objective optimization model is constructed with the maximization of the standard perceived utility value of both parties as the optimization goal, and the solution is obtained.

[0154] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions.

[0155] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0156] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity 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, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0157] The above embodiments should be understood to be only used to illustrate the present invention and not to limit the protection scope of the present invention. After reading the contents of the present invention, technicians can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims

1. A data integrity audit matching method based on psychological perception and preference, characterized in that: The following steps are involved: S1. Calculate the subjective and objective weights ω of each evaluation index s and ω o , and calculate the combined weight ω of each indicator based on the subjective and objective weights of the indicators C , and then use the improved Topsis comprehensive evaluation method to calculate the weight of each audit plan ω a ; S2. Construct a weighted judgment matrix H based on the user preference sets y[M] and y[N][M]. ij According to the weighted judgment matrix, the comprehensive non-fuzzy evaluation matrix E is calculated using the mean area method. ij , using Topsis to calculate the weights of each audit scheme considering user preference requirements b , Gather ω a and ω b The weights of the two are used to obtain the final weight of each audit scheme ω f ; S3. According to the final weight of each audit scheme calculated, based on the preference requirements of both parties, the comprehensive preference value tp of audit tasks to audit resources and audit resources to audit tasks is quantitatively calculated. ij and rp ij , and construct the preference order matrix TPO and RPO, and then convert the preference order matrix into the preference utility value matrix TPU and RPU according to the preference utility function; S4. Using the improved disappointment theory (i.e., the preference order values ​​of both parties are quantified, and the disappointment avoidance effect is considered when constructing the pleasure function, and the subjective characteristics of both parties are considered when calculating the perceived utility value matrix of both parties, rather than completely rational individuals), the disappointment function and the pleasure function form considering the disappointment avoidance effect are respectively constructed, and the comprehensive loss value matrix d(r ij ), d`(t ij ) and the comprehensive happiness value matrix e(r ij ), e`(t ij ), and then modify the preference utility value to obtain the perceived utility value matrix u(r ij ),u`(t ij ); S5. Standardize the perceived utility value matrix of both parties to obtain the standard perceived utility value. and Maximization is taken as the optimization goal, a dual-objective optimization model is constructed, and then solved.

2. According to claim 1, a data integrity audit matching method based on psychological perception and preference is characterized in that: In step S1, the subjective weight and objective weight of the evaluation index and the subjective and objective combined weight are calculated, and then the weight of each audit scheme is calculated based on the improved Topsis comprehensive evaluation method, which specifically includes: S11. Based on the data results of the twelve expert questionnaires obtained from the survey, the improved G1 ordinal relationship analysis method was used to calculate the subjective weights of the evaluation indicators. That is, the opinions of multiple experts were comprehensively considered when using the G1 ordinal relationship analysis method to calculate the subjective weights of the indicators. This is more in line with the objectivity of group decision-making and avoids the deviation that may be caused by the subjective empowerment of a single expert. S12. According to the attribute set of the data integrity audit scheme, the objective weight of the evaluation index is calculated using the improved Critic objective weight method. That is, when using the Critic method to calculate the indicator variability, the influence of the average value on the indicator variability is considered, making the calculation of the coefficient of variation more reasonable and accurate. S13. Use linear weighting method to calculate the subjective and objective combined weight of each indicator. S14. The improved Topsis comprehensive evaluation method is used to calculate the weight of each audit plan. That is, when calculating the weighted Euclidean distance between each audit plan and the positive ideal value and the negative ideal value, the subjective and objective combined weights of the indicators are combined, which effectively improves the rationality and accuracy of the weight calculation results.

3. According to claim 1, a data integrity audit matching method based on psychological perception and preference is characterized in that: In step S2, the weights of each audit scheme considering the user's preference requirements are calculated and the final weights of each audit scheme are obtained by combining the weights of the two, which specifically includes: S21. Based on triangular fuzzy numbers, the user language evaluation is fuzzified and quantified, and a weighted judgment matrix is ​​calculated according to the user preference set to obtain a comprehensive fuzzy evaluation matrix; S22. Use the mean area method to defuzzify the comprehensive fuzzy evaluation matrix to obtain a comprehensive non-fuzzy evaluation matrix, and use the Topsis comprehensive evaluation method to calculate the weights of each audit scheme considering user preference requirements; S23. Using the linear weighting method, the weights of the two are combined to calculate the final weight of each audit plan.

4. According to claim 1, a data integrity audit matching method based on psychological perception and preference is characterized in that: In step S3, the comprehensive preference values ​​of the audit task to the audit resource and the audit resource to the audit task are quantified and the preference utility values ​​of both parties are calculated based on the preference utility function, which specifically includes: S31. Describe the attributes of the user audit task and audit resource, quantify and calculate the preference values ​​of each party according to the preference requirements of both parties, and use the linear weighted method to calculate the comprehensive preference value of both parties, and calculate the preference sequence value according to the comprehensive preference value of the individual; S32. According to the preference order matrix of the matching parties, a preference utility function is constructed, and the preference utility value is calculated to obtain the preference utility value matrix of the audit task to the audit resource and the preference utility value matrix of the audit resource to the audit task.

5. According to claim 1, a data integrity audit matching method based on psychological perception and preference is characterized in that: In step S4, the comprehensive loss value and comprehensive pleasure value matrix of the matching parties are calculated and the preference utility value is corrected to obtain the perceived utility value matrix, which specifically includes: S41. Based on the improved disappointment theory, construct the disappointment function and calculate the subject A i For subject B j Matches instead of B k The loss value of the audit task to the audit resource is d(r ij ) is an arithmetic mean. Similarly, construct a happiness function and calculate the comprehensive happiness value matrix of the matching parties under the consideration of the disappointment avoidance effect. S42. According to the comprehensive pleasure value and comprehensive loss value matrix of the matching parties, the preference utility value matrix of both parties is modified to obtain the perceived utility value matrix of the audit task to the audit resources and the perceived utility value matrix of the audit resources to the audit task.

6. A data integrity audit matching method based on psychological perception and preference according to claim 1, characterized in that: In step S5, the perceived utility value matrix is ​​standardized, and the dual-objective optimization model is constructed and solved with the maximization of the standard perceived utility value as the optimization goal, which specifically includes: S51. Standardize the perceived utility value, take maximizing the standard perceived utility value of both parties as the optimization goal, build a dual-objective optimization model, and use a linear weighted method to convert the dual-objective optimization model into a single-objective optimization model; S52. Use KM algorithm to find the optimal matching result.

7. An electronic device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the user data integrity audit matching method based on psychological perception and preference as claimed in any one of claims 1 to 6 is implemented.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the user data integrity audit matching method based on psychological perception and preference is implemented as described in any one of claims 1 to 6.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the user data integrity audit matching method based on psychological perception and preference is implemented as described in any one of claims 1 to 6.

Images