Analytic hierarchy process (AHP) methods, apparatus, equipment, and storage media

Abstract

This application discloses a hierarchical analysis method, apparatus, device, and storage medium. The method includes: establishing a hierarchical structure model based on the attribute parameters of current candidate solutions; constructing a judgment matrix based on the hierarchical structure model; wherein the elements in the judgment matrix represent the relative importance between the attribute parameters; performing antlion optimization on the judgment matrix to obtain a weight vector and a consistency index; wherein the weight vector represents the relative importance of the current candidate solution in the candidate solution set, and the consistency index represents the rationality of the current candidate solution; and determining a target solution from the candidate solution set based on the weight vector and consistency index of each candidate solution.

Description

Technical Field

[0001] This application relates to information processing technology, including but not limited to hierarchical analysis methods, apparatus, devices, and storage media. Background Technology

[0002] The Analytic Hierarchy Process (AHP) is a commonly used evaluation method when analyzing complex problems. After gaining a deep understanding of the research object and grasping its development laws, decision-makers select indicators to construct a hierarchical structure for analysis. However, solving for the indicators in the hierarchical structure is quite complex and computationally difficult. Summary of the Invention

[0003] In view of this, the hierarchical analysis method, apparatus, device, and storage medium provided in this application utilize the advantages of the antlion optimization method, such as fewer computational parameters, simpler calculation process, and faster calculation speed, which can effectively reduce the difficulty of calculating the judgment matrix weight vector and consistency index in hierarchical analysis. The hierarchical analysis method, apparatus, device, and storage medium provided in this application are implemented as follows:

[0004] The hierarchical analysis method provided in this application includes: establishing a hierarchical structure model based on the attribute parameters of the current candidate solution; constructing a judgment matrix based on the hierarchical structure model; wherein the elements in the judgment matrix are used to characterize the relative importance between the attribute parameters; performing antlion optimization on the judgment matrix to obtain a weight vector and a consistency index; wherein the weight vector is used to characterize the relative importance of the current candidate solution in the candidate solution set, and the consistency index is used to characterize the rationality of the current candidate solution; and determining a target solution from the candidate solution set based on the weight vector and consistency index of each candidate solution.

[0005] The hierarchical analysis apparatus provided in this application includes: a construction module, configured to establish a hierarchical structure model based on the attribute parameters of the current candidate scheme; the construction module is further configured to construct a judgment matrix based on the hierarchical structure model; wherein the elements in the judgment matrix are used to characterize the relative importance between the attribute parameters; a processing module, configured to perform antlion optimization processing on the judgment matrix to obtain a weight vector and a consistency index; wherein the weight vector is used to characterize the relative importance of the current candidate scheme in the candidate scheme set, and the consistency index is used to characterize the rationality of the current candidate scheme; and a determination module, configured to determine a target scheme from the candidate scheme set based on the weight vector and consistency index of each candidate scheme.

[0006] The electronic device provided in this application includes a memory and a processor. The memory stores a computer program that can run on the processor. When the processor executes the program, it implements the method described in this application.

[0007] The computer-readable storage medium provided in this application embodiment stores a computer program thereon, which, when executed by a processor, implements the method provided in this application embodiment.

[0008] In this embodiment, a hierarchical analysis method is provided. This method involves establishing a hierarchical structure model; constructing a judgment matrix based on the hierarchical structure model; then performing antlion optimization on the matrix to obtain a weight vector and a consistency index; finally, determining the target scheme from the candidate scheme set based on the weight vector and consistency index of each candidate scheme. Thus, by applying the antlion optimization method to solve for the weight vector and consistency index in hierarchical analysis, the advantages of the antlion optimization method—few computational parameters, simple computation process, and fast computation speed—are utilized, effectively reducing the difficulty of calculating the weight vector and consistency index of the judgment matrix in hierarchical analysis. Attached Figure Description

[0009] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with this application and, together with the specification, serve to explain the technical solutions of this application.

[0010] Figure 1 A schematic diagram illustrating the implementation flow of a hierarchical analysis method provided in this application embodiment;

[0011] Figure 2 A schematic diagram illustrating the implementation flow of a hierarchical analysis method provided in this application embodiment;

[0012] Figure 3 A schematic diagram illustrating the implementation flow of a hierarchical analysis method provided in this application embodiment;

[0013] Figure 4 A schematic diagram illustrating the implementation flow of a hierarchical analysis method provided in this application embodiment;

[0014] Figure 5 This is a flowchart illustrating the improved antlion optimization algorithm according to an embodiment of this application.

[0015] Figure 6 This is a schematic diagram of the structure of the hierarchical analysis device according to an embodiment of this application;

[0016] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the specific technical solutions of this application will be further described in detail below with reference to the accompanying drawings of the embodiments of this application. The following embodiments are used to illustrate this application, but are not intended to limit the scope of this application.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0019] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0020] It should be noted that the terms "first, second, third" used in the embodiments of this application are used to distinguish similar or different objects and do not represent a specific order of objects. It can be understood that "first, second, third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0021] This application provides a hierarchical analysis method applied to an electronic device. This electronic device can be various types of computer devices with information processing capabilities, such as desktop computers, laptops, or handheld computers. It can also be a mobile terminal, such as a mobile phone, in-vehicle computer, tablet computer, or POS machine. The functions implemented by this method can be achieved by a processor in the electronic device calling program code. The program code can be stored in a computer storage medium. Therefore, the electronic device includes at least a processor and a storage medium.

[0022] Figure 1 This is a schematic diagram illustrating the implementation flow of the hierarchical analysis method provided in the embodiments of this application, as shown below. Figure 1 As shown, the method may include the following steps 101 to 104:

[0023] Step 101: Establish a hierarchical structure model based on the attribute parameters of the current candidate solutions.

[0024] The Analytic Hierarchy Process (AHP) is a systems analysis method that decomposes a complex problem layer by layer according to the logical relationships of its attributes, forming a hierarchical structure model for analysis. This reduces the decision-making difficulty of the problem and provides solutions to complex problems. After gaining a deep understanding of the research object and grasping its development laws, decision-makers select attribute parameters to construct a hierarchical structure for analysis. They then use mathematical methods to quantify these attribute parameters and make judgments about the evaluated objects based on the calculation results.

[0025] In the embodiments of this application, the attribute parameters of the candidate solutions are predetermined. When determining the attribute parameters of the candidate solutions, they can be selected according to the specific application scenario of the solution, that is, according to actual needs.

[0026] This method can be applied in various scenarios. For example, users may want to choose the most suitable travel destination from multiple candidate tourist destinations; in an examination system, teachers may want to choose the most reasonable set of test papers from multiple sets of test papers; and users may want to choose the most suitable job from multiple job opportunities.

[0027] Let's take the selection of a tourist destination as an example: A user wants to choose a target tourist destination from three candidate tourist destinations. Let's assume these three destinations are P1 (Suzhou), P2 (Hangzhou), and P3 (Nanjing). The attribute parameters that influence the user's choice of destination are five factors: B1 (scenery), B2 (cost), B3 (accommodation), B4 (food), and B5 (trip). Based on these five attribute parameters of the candidate destinations, a hierarchical model can be established, including a target layer, a criterion layer, and a destination layer.

[0028] Step 102: Construct a judgment matrix based on the hierarchical structure model; wherein the elements in the judgment matrix are used to characterize the relative importance between attribute parameters.

[0029] After establishing the hierarchical model, a judgment matrix is ​​constructed based on the influence parameters of candidate solutions within the model. The judgment matrix represents the relative importance of each attribute parameter at each level of the hierarchical model. Relative importance refers to whether one attribute parameter is more important than another when constructing the judgment matrix; this value is predetermined based on the specific application scenario. For example, when choosing a tourist destination, if the scenery attribute is considered slightly more important than the cost attribute, the ratio of scenery to cost in the judgment matrix is ​​set to 2; if the travel attribute is considered significantly more important than the accommodation attribute, the ratio of travel to accommodation is set to 5. Based on this criterion, the ratios between the attribute parameters of each candidate solution are determined pairwise, ultimately constructing the judgment matrix.

[0030] Step 103: Perform antlion optimization on the judgment matrix to obtain a weight vector and a consistency index; whereby the weight vector is used to characterize the relative importance of the current candidate solution in the candidate solution set, and the consistency index is used to characterize the rationality of the current candidate solution.

[0031] In the analytic hierarchy process (AHP), after constructing the decision matrix, it is necessary to calculate the weight vector and consistency index of the decision matrix, and evaluate candidate solutions based on these two parameters. However, the calculation process is quite complex and difficult.

[0032] Antlion optimization is a swarm intelligence optimization algorithm that simulates the behavior of antlion larvae preying on ants. Its optimization ability mainly depends on the interaction and influence between individual ants and individual antlions.

[0033] In this embodiment, based on the optimization capability of the antlion optimization method, it is applied to the hierarchical analysis method to identify elite individuals. The spatial position of the elite individuals is used as the weight vector, and the fitness function value of the elite individuals is used as the consistency index. It is evident that this method of applying the antlion optimization algorithm to hierarchical analysis leverages the advantages of the antlion optimization algorithm—few computational parameters, simple calculation process, and fast calculation speed—thereby effectively reducing the difficulty of calculating the judgment matrix weight vector and consistency index in hierarchical analysis.

[0034] Step 104: Determine the target solution from the candidate solution set based on the weight vector and consistency index of each candidate solution.

[0035] Taking tourist destination selection as an example, after performing Antlion optimization on the constructed judgment matrix, assuming the weight vectors for Suzhou, Hangzhou, and Nanjing among the candidate options are 0.105, 0.637, and 0.258 respectively, it can be seen that Hangzhou has the largest weight value among the three candidate options. Therefore, Hangzhou is likely to be the target option. However, when calculating the weight vector of this option, the values ​​of the elements in the judgment matrix may not be entirely reasonable. Therefore, it is necessary to continue to judge the consistency index of the candidate option Hangzhou. When it is determined that the consistency index of this option meets the consistency requirements, that is, when the values ​​of the elements in the judgment matrix of the current option are reasonably set, Hangzhou is determined as the target option.

[0036] Figure 2 This is a schematic diagram illustrating the implementation flow of the hierarchical analysis method provided in the embodiments of this application, as shown below. Figure 2 As shown, the method may include the following steps 201 to 207:

[0037] Step 201: Establish a hierarchical structure model based on the attribute parameters of the current candidate solutions;

[0038] Step 202: Construct a judgment matrix based on the hierarchical structure model;

[0039] Step 203: Generate antlion populations and ant populations based on the judgment matrix; wherein the population size of the antlion population and the population size of the ant population have the same dimension as the judgment matrix.

[0040] In this embodiment, after constructing the judgment matrix, an initial antlion population and an initial ant population are generated based on the elements in the judgment matrix. For example, the constructed judgment matrix is ​​A = (a ij ) N×N The judgment matrix has N rows (horizontal dimension) and N columns (vertical dimension). Based on the judgment matrix, the population sizes of antlion colonies and ant colonies are set, with each population size being N. In other words, the population sizes of antlion colonies and ant colonies have the same dimensions as the judgment matrix.

[0041] Step 204: Determine the spatial location and fitness function value of each individual in the antlion population and the ant population; wherein, the fitness function value is used to characterize the degree of fitness of an individual in the population.

[0042] After obtaining the antlion population and the ant population, the spatial position of each individual in the antlion population and the ant population is first initialized, initializing N antlions: M antlion_i =(al1,al2,...,al n ), i = 1, 2, ..., N, N ants: M ant_i =(a1,a2,...,a n ), i = 1, 2, ..., N. Where N is the population size, al i and a i Each component is a uniformly random number on [0, 1], thus obtaining the spatial location of each individual in the initial ant colony and the initial antlion colony.

[0043] The fitness function value is used to measure the quality of an individual in a population. In other words, it measures the degree of fitness of an individual in the population. Individuals with high fitness are considered excellent individuals, while individuals with low fitness are considered poor individuals.

[0044] Step 205: Select elite individuals from the antlion population and the ant population based on the fitness function value of each individual.

[0045] In some embodiments, this can be achieved by performing the following steps 2051 to 2056:

[0046] Step 2051: Individuals in the antlion population whose fitness function values ​​meet the candidate criteria are identified as candidate individuals.

[0047] Candidate criteria can be either the minimum or maximum fitness function value. For example, the fitness function values ​​of each individual in the antlion population can be sorted in ascending order, and the individual with the minimum fitness function value in the sorted results is selected as a candidate individual. Alternatively, the fitness function values ​​can be sorted in ascending order, and the individual with the maximum fitness function value in the sorted results is selected as a candidate individual. Another example is sorting the fitness function values ​​in descending order, and the individual with the minimum fitness function value in the sorted results is selected as a candidate individual. Yet another example is sorting the fitness function values ​​in descending order, and the individual with the maximum fitness function value in the sorted results is selected as a candidate individual. These selected candidate individuals are the initial candidate individuals, i.e., the elite individuals in the initial antlion population.

[0048] Step 2052: Update the ant population based on the spatial location of the candidate individuals to obtain the updated ant population.

[0049] In the antlion optimization algorithm, individual ants wander randomly, and their positions are constantly changing. Therefore, after determining the initial candidate individuals, it is necessary to update the positions of the ants in the ant colony based on the spatial location of the initial candidate individuals to obtain the updated ant colony.

[0050] In some embodiments, the spatial position of each individual in the ant colony is updated based on the spatial position of the candidate individual; and the fitness function value of the corresponding individual in the ant colony is updated based on the updated spatial position of the individual in the ant colony.

[0051] Understandably, the fitness function value of each individual ant will change after its spatial location changes. Therefore, after obtaining the updated spatial location of each individual ant in the ant colony, it is necessary to calculate its corresponding fitness function value based on the individual ant's spatial location. This fitness function value will also be different from the fitness function value before the update.

[0052] Step 2053: Mix the updated ant population with the antlion population to obtain a mixed population.

[0053] After an individual's spatial location is updated in an ant colony, its fitness function value will also change accordingly. Therefore, the initial candidate individuals may not necessarily be individuals that meet the candidate conditions. Thus, it is necessary to mix the updated ant colony with the initial antlion colony to obtain a mixed colony. In the mixed colony, the candidate individuals that currently meet the candidate conditions are selected based on the updated fitness function value of each individual.

[0054] Step 2054: Update the first individual in the mixed population whose fitness function value satisfies the candidate condition to obtain a new candidate individual.

[0055] The method for selecting candidate individuals in the mixed population is the same as that for selecting candidate individuals in the initial antlion population, and will not be repeated here.

[0056] Step 2055: Determine whether the number of updates meets the first convergence condition; if not, return to step 2052 and continue to update the current ant population according to the spatial location of the new candidate individuals to obtain new candidate individuals; if the condition is met, proceed to step 2056.

[0057] Understandably, in the process of antlion optimization, it is far from enough to optimize the population only once (update once) in order to find the best candidate individual (i.e. the final elite individual). Multiple optimizations are required until the number of optimizations meets the convergence condition. The candidate individual in the mixed population obtained from the final optimization can be determined as the elite individual.

[0058] In this embodiment, determining whether the number of updates meets the first convergence condition is equivalent to determining whether the number of updates has reached the maximum number of iterations. For example, the maximum number of iterations can be set to 1000; however, the specific number can be set according to user needs and is not limited thereto.

[0059] The update count refers to the number of times the ant population is updated. When it is determined that the update count does not meet the first convergence condition, the process returns to step 2052, that is, based on the spatial position of the new candidate individuals obtained in this update, the spatial position of each individual in the ant population is updated to obtain the updated ant population again, and new candidate individuals are selected from the mixed population again. When the update count meets the maximum number of iterations, step 2056 is executed, and based on the mixed population obtained in the last update, the candidate individuals obtained in the last update are determined and regarded as elite individuals.

[0060] Step 2056: The new candidate individuals obtained based on the current mixed population are identified as elite individuals.

[0061] Step 206: Determine the spatial location of elite individuals as a weight vector, and determine the fitness function value of elite individuals as a consistency index.

[0062] In this embodiment, the calculation of the weight vector in the judgment matrix is ​​transformed into a minimum optimization problem; and through a pre-established mathematical model, the calculation of the consistency index is transformed into the calculation of the fitness function value in the antlion optimization algorithm. The core idea of ​​this method is: based on the antlion optimization algorithm, elite individuals in the population are identified, the spatial position of the elite individuals is used as the weight vector, and the fitness function value of the elite individuals is used as the consistency index; thus, the relatively complex problem of calculating the weight vector and consistency index is transformed into an optimization problem, effectively utilizing the advantages of the antlion optimization algorithm, such as fewer computational parameters, simple calculation process, and fast calculation speed, thereby reducing the difficulty of calculating the weight vector of the judgment matrix and the consistency index in the hierarchical analysis.

[0063] Step 207: Determine the target solution from the candidate solution set based on the weight vector and consistency index of each candidate solution.

[0064] Figure 3 This is a schematic diagram illustrating the implementation flow of the hierarchical analysis method provided in the embodiments of this application, as shown below. Figure 3 As shown, the method may include the following steps 301 to 318:

[0065] Step 301: Establish a hierarchical structure model based on the attribute parameters of the current candidate solutions;

[0066] Step 302: Construct a judgment matrix based on the hierarchical structure model;

[0067] Step 303: Generate antlion population and ant population based on the judgment matrix;

[0068] Step 304: Determine the spatial location and fitness function value of each individual in the antlion population and the ant population;

[0069] Step 305: Individuals in the antlion population whose fitness function values ​​meet the candidate criteria are identified as candidate individuals;

[0070] Step 306: Update the ant population based on the spatial location of the candidate individuals to obtain the updated ant population;

[0071] Step 307: Mix the updated ant population with the antlion population to obtain a mixed population;

[0072] Step 308: Perform chaotic optimization on the spatial position of the first individual in the mixed population whose fitness function value satisfies the candidate condition to obtain the spatial position of the second individual.

[0073] In the antlion optimization algorithm, if the selected elite individual is not the global optimum within its local optimum neighborhood, it can easily lead the entire population into a local optimum, reducing population diversity. To address this issue, in this embodiment, when candidate individuals are identified from the mixed population, a chaotic optimization method is introduced to optimize these candidate individuals. In the chaotic optimization process, by searching for superior individuals in the vicinity of the candidate individual to replace the currently selected one, the population can avoid falling into local optima and improve population diversity.

[0074] In some embodiments, the chaotic optimization processing of the spatial position of the first body can be achieved by mapping the spatial position of the first body from the first space to the second space to obtain the mapped position of the second body; and reversing the mapped position of the second body from the second space to the first space to obtain the spatial position of the second body; wherein the spatial position of the second body is different from the spatial position of the first body.

[0075] It should be noted that the mapping function used when mapping the spatial position of the first individual is different from the inverse mapping function used when mapping the position of the second individual. That is, the spatial position of the second individual is different from that of the first individual; the second individual is a new individual generated after mapping the first individual. Accordingly, the fitness function value of the second individual is also different from that of the first individual.

[0076] Step 309: Determine the fitness function value of the second individual based on its spatial location;

[0077] Step 310: Based on the fitness function value of the second individual and the fitness function value of the first individual, determine whether the second individual meets the update condition; if the second individual meets the update condition, proceed to step 311; otherwise, proceed to step 312.

[0078] In this embodiment of the application, the update condition is related to the candidate condition.

[0079] When the candidate condition is the minimum fitness function value, the update condition is that the fitness function value of the second person is less than the fitness function value of the first person. That is, if the fitness function value of the second person is less than the fitness function value of the first person, then step 311 is executed to update the second person to the first person; if the fitness function value of the second person is greater than or equal to the fitness function value of the first person, then step 312 is executed to retain the first person.

[0080] When the candidate condition is that the fitness function value is the largest, the update condition is that the fitness function value of the second person is greater than the fitness function value of the first person. That is, if the fitness function value of the second person is greater than the fitness function value of the first person, then step 311 is executed to update the second person to the first person; if the fitness function value of the second person is less than or equal to the fitness function value of the first person, then step 312 is executed to retain the first person.

[0081] Step 311: Update the second body to the first body;

[0082] Step 312, retain the first body;

[0083] Step 313: Determine whether the number of optimization iterations meets the second convergence condition. If the number of optimization iterations does not meet the second convergence condition, return to step 308 and continue to perform chaotic optimization processing on the spatial position of the first individual; otherwise, execute step 314 and determine the first individual obtained so far as a new candidate individual.

[0084] Understandably, when performing chaotic optimization on the spatial position of the first body, processing it only once is far from enough. Therefore, multiple optimization processes are required until the number of processes meets the convergence condition.

[0085] In this embodiment, determining whether the number of optimization attempts meets the second convergence condition is equivalent to determining whether the maximum number of optimization attempts has been reached. For example, the maximum number of attempts can be set to 200; however, the specific number can be set according to user needs and is not limited thereto.

[0086] Step 314: The first individual obtained so far is identified as a new candidate individual;

[0087] Step 315: Determine whether the number of updates meets the first convergence condition; if not, return to step 306 and continue to update the current ant population according to the spatial location of the new candidate individuals to obtain new candidate individuals; if the number of updates meets the first convergence condition, proceed to step 316.

[0088] Step 316: Identify the new candidate individuals obtained based on the current mixed population as elite individuals;

[0089] Step 317: Determine the spatial location of elite individuals as a weight vector, and determine the fitness function value of elite individuals as a consistency index.

[0090] Step 318: Determine the target solution from the candidate solution set based on the weight vector and consistency index of each candidate solution.

[0091] Figure 4This is a schematic diagram illustrating the implementation flow of the hierarchical analysis method provided in the embodiments of this application, as shown below. Figure 4 As shown, the method may include the following steps 401 to 415:

[0092] Step 401: Establish a hierarchical structure model based on the attribute parameters of the current candidate solutions;

[0093] Step 402: Construct a judgment matrix based on the hierarchical structure model;

[0094] Step 403: Generate antlion population and ant population based on the judgment matrix;

[0095] Step 404: Determine the spatial location and fitness function value of the ant lion population and each individual in the ant population;

[0096] Step 405: Individuals in the antlion population whose fitness function values ​​meet the candidate criteria are identified as candidate individuals;

[0097] Step 406: Update the ant population based on the spatial location of the candidate individuals to obtain the updated ant population;

[0098] Step 407: Mix the updated ant population with the antlion population to obtain a mixed population;

[0099] Step 408: Update the first individual in the mixed population whose fitness function value satisfies the candidate condition to obtain a new candidate individual;

[0100] Step 409: Identify undesirable individuals in the mixed population.

[0101] Understandably, unfit individuals refer to those with poor fitness within a population, individuals that cannot adapt well to the population's evolutionary direction. In mixed populations, the presence of unfit individuals slows down the convergence speed and accuracy of the antlion optimization algorithm, and may even lead to evolutionary stagnation. To address these issues, in this embodiment, after identifying unfit individuals in the mixed population, directional mutation processing is performed on these individuals to remove the uncertainty caused by random directional mutations. This ensures that the mutation direction of the unfit individuals' spatial positions points towards the spatial positions of elite individuals, thereby accelerating the algorithm's convergence speed and improving its solution accuracy.

[0102] In this embodiment, the criteria for determining which individuals in the mixed population are undesirable individuals are also related to candidate criteria. When the candidate criterion is the minimum fitness function value, individuals in the mixed population with fitness function values ​​greater than a certain threshold are identified as undesirable individuals; when the candidate criterion is the maximum fitness function value, individuals in the mixed population with fitness function values ​​less than a certain threshold are identified as undesirable individuals.

[0103] In some embodiments, for ease of algorithm implementation, when the candidate condition is the minimum fitness function value, the last n individuals in the mixed population can be identified as undesirable individuals; when the candidate condition is the maximum fitness function value, the first n individuals in the mixed population can be identified as undesirable individuals. Here, n is any value, such as 1 / 3 or 1 / 2 of the population size in the mixed population, and is not limited thereto.

[0104] Step 410: The spatial location of the undesirable individuals in the mixed population is mutated so that the spatial location of the undesirable individuals is within a specific range of the spatial location of the new candidate individuals, thereby obtaining the updated mixed population.

[0105] The purpose of modifying the spatial location of undesirable individuals is to make their modified locations more similar to those of elite individuals (i.e., new candidate individuals), that is, to make the modified spatial location of undesirable individuals closer to the spatial location of elite individuals.

[0106] In some embodiments, the spatial location of the defective individual can be mutated by performing beta mutation. Beta mutation can achieve a better perturbation effect and is beneficial for the algorithm to control the direction and location boundary of the mutation.

[0107] In some embodiments, the variation processing of the spatial location of undesirable individuals in a mixed population can be achieved by performing the following steps 4101 to 4103:

[0108] Step 4101: Determine the probability that the spatial location of the defective individual is within a specific range of the spatial location of the new candidate individual.

[0109] The specific range of the spatial location of a new candidate individual refers to the range that is as close as possible to the location of the candidate individual. This range can be set according to user needs and is not limited in this respect.

[0110] It should be noted that the probability is set in the interval [0, 1]. If the probability is close to 1, it means that the position of the problematic individual is very close to the position of the candidate individual; if the probability is close to 0, it means that the position of the problematic individual is far away from the position of the candidate individual.

[0111] Step 4102: Determine the difference between the spatial location of the defective individual and the spatial location of the new candidate individual;

[0112] Step 4103: Update the spatial location of the bad individuals based on the probability and the difference.

[0113] In the embodiments of this application, after determining the probability σ and the difference (x) id -x eliteAfter that, the spatial location of the defective individual can be updated according to the following formula 1. Where, x id x represents the spatial location of a delinquent individual. elite This indicates the spatial location of the new candidate individual.

[0114] x′ id =x elite +σ(x id -x elite ) (Formula 1);

[0115] Understandably, the probability σ lies in the interval [0, 1]. If σ is close to 1, it means that the position of the problematic individual is very close to the position of the candidate individual. Correspondingly, the difference (x) id -x elite The product of probability σ and difference is small, so the product of probability σ and difference is also small, resulting in a smaller updated value for the undesirable individual x′. id Spatial location close to candidate individual x elite Spatial location; if σ is close to 0, it indicates that the location of the defective individual is far from the location of the candidate individual, and correspondingly, the difference (x) id -x elite Similarly, after multiplying the probability σ by the difference, the product is still very small, thus making the updated undesirable individual x... i ' d Spatial location close to candidate individual x elite Therefore, by performing Beta mutation on undesirable individuals, the spatial location of undesirable individuals can be made to tend towards the spatial location of elite individuals (i.e., candidate individuals), thereby accelerating the convergence speed of the algorithm and improving the solution accuracy.

[0116] Step 411: Divide the updated mixed population into a first population and a second population, and update the first population to an antlion population and the second population to an ant population.

[0117] After replacing the unfit individuals in the mixed population with better individuals that are closer to the elite individuals, the updated mixed population is re-divided into two populations: the better-fit individuals in the mixed population are classified as the antlion population, and the poorer-fit individuals in the mixed population are classified as the ant population.

[0118] Step 412: Determine whether the number of updates meets the first convergence condition; if not, return to step 406 and continue to update the current ant population according to the spatial location of the new candidate individuals to obtain new candidate individuals; if the condition is met, proceed to step 413.

[0119] Step 413: Identify the new candidate individuals obtained based on the current mixed population as elite individuals;

[0120] Step 414: Determine the spatial location of elite individuals as a weight vector, and determine the fitness function value of elite individuals as a consistency index.

[0121] Step 415: Determine the target solution from the candidate solution set based on the weight vector and consistency index of each candidate solution.

[0122] The Analytic Hierarchy Process (AHP) is a commonly used evaluation method when analyzing complex problems. After gaining a deep understanding of the research object and grasping its development patterns, decision-makers select indicators to construct a hierarchical structure for analysis. They then use mathematical methods to quantify these indicators and make judgments based on the calculation results. Since the indicator calculation problem in the AHP can be transformed into an optimization problem, the following swarm intelligence algorithms are currently the main technologies used:

[0123] (1) The calculation of the weight vector in the judgment matrix is ​​transformed into a minimum optimization problem by using the Accelerating Genetic Algorithm-Analytic Hierarchy Process (AGA-AHP) to verify the consistency of the judgment matrix and calculate its weight vector at the same time.

[0124] (2) The Particle Swarm Optimization-Analytic Hierarchy Process (PSO-AHP) is used to calculate the weights of each element with high consistency index accuracy.

[0125] (3) Improve the calculation of the ranking weights of the judgment matrix and its consistency test by using the improved distribution estimation algorithm (Particle Filter-Estimation of distribution algorithms, PF-EDA2).

[0126] Analysis of the three methods above reveals the following problems:

[0127] (1) The AGA-AHP algorithm can destroy the structure due to the abruptness of crossover and mutation operations, which affects the accuracy of the results;

[0128] (2) Due to the influence of parameters, the PSO-AHP algorithm may suffer from problems such as premature convergence, slow convergence speed and low stability.

[0129] (3) The PF-EDA2 algorithm needs to calculate the covariance matrix of each generation and obtain matrix decomposition (Cholesky factorization, Cholesky) during the calculation process, which is quite complicated.

[0130] To address the aforementioned issues, a method is proposed to improve the basic antlion optimization algorithm using chaotic optimization and Beta mutation strategies. This method is then applied to the calculation of the judgment matrix weight vector in hierarchical analysis, thus comprehensively considering the problems mentioned above.

[0131] Based on this, the exemplary application of the embodiments of this application in a practical application scenario will be described below.

[0132] In this application embodiment, a chaotic antlion hierarchical analysis algorithm is proposed. This method improves the basic antlion optimization algorithm by using chaotic optimization strategy and Beta mutation strategy, and applies it to the calculation of the judgment matrix weight vector in hierarchical analysis. It can effectively reduce the difficulty of the judgment matrix weight vector calculation problem, and has the advantages of high accuracy and good stability.

[0133] The antlion optimization algorithm is a swarm intelligence optimization algorithm that simulates the predatory behavior of antlion larvae. Its optimization ability mainly relies on the interaction and influence between individual ants and antlions. If the selected elite individual is not the global optimum within its local optimum neighborhood, it can easily lead the entire population into a local optimum, reducing population diversity. At the same time, individuals with poor fitness can slow down the algorithm's convergence speed, reduce convergence accuracy, and even cause evolutionary stagnation.

[0134] To improve population diversity and allow elite individuals to escape the constraints of local optima and approach the global optimum within the local optimal neighborhood, this embodiment introduces a logically self-mapping chaotic sequence to optimize elite individuals in the basic antlion optimization algorithm. Its mathematical expression is as follows:

[0135]

[0136] The process of chaos optimization is shown in steps 1 to 3 below:

[0137] Step 1: At some point during the search process, the elite antlion i is located in D-dimensional space at position x. id First, map each dimension of the antlion's spatial location to the interval [-1, 1] according to Formula 3:

[0138]

[0139] Among them, a id Let b represent the lower bound of the d-th dimension variable of the elite antlion i. id Let represent the upper bound of the d-th dimension variable of the elite antlion i.

[0140] Step 2, use Formula 2 to transform the variable y idPerform carrier operations to obtain a new chaotic variable sequence y′ id Transform back to the original solution space according to Formula 4:

[0141]

[0142] Step 3: During this process, if a better solution is found, the original position of the elite antlion is updated to the current better position; otherwise, the chaotic search continues until the maximum number of searches is reached, at which point the search stops.

[0143] In the antlion optimization algorithm, targeted mutation is used for individuals with poor fitness values ​​to eliminate the uncertainty caused by random mutation. The mutation direction is directed towards elite individuals, which can accelerate the convergence speed and improve the solution accuracy. In this embodiment, a Beta mutation strategy is used to mutate individuals in the population.

[0144] The Beta distribution is a set of continuous probability distributions defined on the interval [0, 1]. The probability density function of the Beta distribution is:

[0145]

[0146] Beta mutation can achieve good perturbation effect and is conducive to the algorithm controlling the mutation direction and position boundary. In this embodiment, an elite-oriented Beta mutation operator is designed according to the Beta distribution for mutation in each iteration cycle. The position mutation formula of elite-oriented Beta mutation is shown in Formula 1 in the above embodiment.

[0147] Figure 5 A flowchart of the improved antlion optimization algorithm is given, such as... Figure 5 As shown, perform the following steps 501 to 512:

[0148] Step 501: Initialize the number, position, maximum number of iterations, fitness function, dimension, and boundary of ants and antlions; where the population size of both ants and antlions is N.

[0149] Step 502: Calculate and sort the fitness function values ​​of ants and antlions, and store the elite antlions;

[0150] Step 503: Set the iteration period t, and update t = t + 1 after each iteration.

[0151] Step 504: Traverse the i-th individual in the ant colony and antlion colony to determine whether the condition i < 2N is met. If i < 2N, proceed to step 505; otherwise, proceed to step 506.

[0152] Step 505: Update ant location;

[0153] Step 506, let i = i + 1;

[0154] Step 507: Calculate the fitness function value of each individual in the ant colony and antlion colony, and sort them.

[0155] Step 508: Apply a logical self-mapping chaotic optimization strategy (i.e., chaotic optimization processing) to the top n elite individuals in the population;

[0156] Step 509: Determine whether the maximum number of chaotic searches has been reached; if the maximum number of chaotic searches has been reached, proceed to step 510; otherwise, return to step 508.

[0157] Step 510: Apply a Beta mutation strategy to the last n individuals with poor fitness.

[0158] Step 511: Calculate and sort the fitness function value of each individual in the ant population and antlion population, and update the antlion positions and elite antlions.

[0159] Step 512: Determine whether the maximum number of iterations has been reached. If the maximum number of iterations has been reached, end the Antlion optimization algorithm; otherwise, return to step 503.

[0160] The modeling of the analytic hierarchy process (AHP) mainly includes five basic steps: (1) establishing a hierarchical structure model; (2) constructing a judgment matrix; (3) hierarchical single sorting; (4) consistency check of the judgment matrix; and (5) hierarchical overall sorting. Among these, solving for the weight vector in the hierarchical single sorting is the key to the algorithm. In this embodiment, the calculation of the weight vector of the judgment matrix is ​​described as the following minimum optimization problem, and a mathematical model of the weight vector is established as shown in Formula 6:

[0161]

[0162] The variable to be optimized is ω. k CIF(n,ω) is a nonlinear optimization function, also known as the consistency index function.

[0163] The chaotic antlion hierarchical analysis method is shown in steps 1 through 7 below:

[0164] Step 1, given a judgment matrix A = (a ij ) n×n Initialize N antlions: M antlion_i =(al1,al2,...,al n ), i = 1, 2, ..., N, N ants: M ant_i =(a1,a2,...,a n ), i = 1, 2, ..., N, where N is the population size, al i and ai Each component is a uniformly random number in [0, 1], and the maximum number of iterations is initialized.

[0165] Step 2, for the antlion M antlion_i For i = 1, 2, ..., N, calculate CI according to Formula 6. i =CIF(n,M) antlion_i ), as M antlion_i The fitness function value;

[0166] Step 3: Sort the fitness function values ​​of each antlion in the group from smallest to largest, and select the antlion with the smallest fitness function value as the current elite antlion.

[0167] Step 4: Update the ant's position and calculate the fitness function value;

[0168] Step 5: Optimize the top n individuals with high fitness values ​​using a chaotic optimization strategy, and generate new individuals using a Beta mutation strategy for the bottom n individuals with low fitness values. Record the current elite antlion individuals.

[0169] Step 6: If the algorithm reaches the maximum number of iterations of the Antlion optimization algorithm (i.e., an example of the first convergence condition), then stop the iteration and jump to step 7; otherwise, jump to step 4 for the next iteration.

[0170] Step 7, Output the spatial location M of the elite antlion individual. elite =(al1,al2,...,al n ), which is the weight vector; the fitness function value CI = CIF(n,M) corresponding to an elite antlion individual. elite The consistency index is reached, and the algorithm ends.

[0171] In this application embodiment, a chaotic antlion hierarchical analysis method is proposed, in which the following improvements are made: (1) introducing a logical self-mapping chaotic sequence to optimize elite antlion individuals; (2) designing a Beta mutation operator to mutate individuals with poor fitness in the population; (3) calculating the judgment matrix weight vector in the hierarchical analysis method using the improved antlion optimization algorithm.

[0172] The chaotic antlion hierarchical analysis algorithm proposed in this application is characterized by: (1) increasing the diversity of the population, enabling elite individuals to escape the constraints of local extremes and move closer to the global optimum in the local optimal neighborhood; (2) performing directional mutation on individuals with poor fitness, enabling the algorithm to effectively jump out of local extremes, accelerate the convergence speed, and have good stability; (3) the algorithm can calculate the weight of each element with high consistency index accuracy, simplifying the solution process of the weight vector.

[0173] Based on the foregoing embodiments, this application provides a hierarchical analysis device, which includes various modules and units included in each module, and can be implemented by a processor; of course, it can also be implemented by specific logic circuits; in the implementation process, the processor can be a central processing unit (CPU), microprocessor (MPU), digital signal processor (DSP) or field programmable gate array (FPGA), etc.

[0174] Figure 6 This is a schematic diagram of the structure of the hierarchical analysis device according to an embodiment of this application, such as... Figure 6 As shown, the device 600 includes a construction module 601, a processing module 602, and a determination module 603, wherein:

[0175] Module 601 is used to build a hierarchical model based on the attribute parameters of the current candidate schemes;

[0176] The construction module 601 is further configured to construct a judgment matrix based on the hierarchical structure model; wherein the elements in the judgment matrix are used to characterize the relative importance between the attribute parameters;

[0177] The processing module 602 is used to perform antlion optimization processing on the judgment matrix to obtain a weight vector and a consistency index; wherein, the weight vector is used to characterize the relative importance of the current candidate solution in the candidate solution set, and the consistency index is used to characterize the rationality of the current candidate solution;

[0178] The determination module 603 is used to determine the target solution from the set of candidate solutions based on the weight vector and consistency index of each candidate solution.

[0179] In some embodiments, the hierarchical analysis device 600 further includes a generation module and a selection module. The generation module is used to generate an antlion population and an ant population based on the judgment matrix. The determination module is used to determine the spatial location and fitness function value of each individual in the antlion population and the ant population. The selection module is used to select elite individuals from the antlion population and the ant population based on the fitness function value of each individual. The determination module is also used to determine the spatial location of the elite individuals as the weight vector and the fitness function value of the elite individuals as the consistency index.

[0180] In some embodiments, the hierarchical analysis device 600 further includes an update module and a mixing module. The determination module is used to determine individuals in the antlion population whose fitness function values ​​satisfy the candidate conditions as candidate individuals. The update module is used to update the ant population according to the spatial location of the candidate individuals to obtain an updated ant population. The mixing module is used to mix the updated ant population with the antlion population to obtain a mixed population. The update module is also used to update the first individual in the mixed population whose fitness function value satisfies the candidate conditions to obtain a new candidate individual. The determination module is also used to continue to update the current ant population according to the spatial location of the new candidate individuals to obtain new candidate individuals, until the number of updates meets the first convergence condition, and then determine the new candidate individuals obtained based on the current mixed population as the elite individuals.

[0181] In some embodiments, the updating module is configured to update the spatial position of each individual in the ant colony based on the spatial position of the candidate individual; and update the fitness function value of the corresponding individual in the ant colony based on the updated spatial position of the individual in the ant colony.

[0182] In some embodiments, the processing module is configured to perform chaotic optimization processing on the spatial position of the first individual to obtain the spatial position of the second individual; the determining module is configured to determine the fitness function value of the second individual based on the spatial position of the second individual; the determining module is further configured to determine whether the second individual meets the update condition based on the fitness function value of the second individual and the fitness function value of the first individual; if the second individual meets the update condition, the second individual is updated to the first individual; if the second individual does not meet the update condition, the first individual is retained; the determining module is further configured to continue performing chaotic optimization processing on the spatial position of the first individual until the number of optimizations meets the second convergence condition, and determine the currently obtained first individual as the new candidate individual.

[0183] In some embodiments, the hierarchical analysis device 600 further includes a mapping module, which is used to map the spatial position of the first individual from a first space to a second space to obtain the mapping position of the second individual; the mapping module is also used to reverse map the mapping position of the second individual from the second space to the first space to obtain the spatial position of the second individual.

[0184] In some embodiments, the hierarchical analysis device 600 further includes a partitioning module; the determining module is used to determine undesirable individuals in the mixed population; the processing module is used to perform mutation processing on the spatial positions of undesirable individuals in the mixed population so that the spatial positions of the undesirable individuals are within a specific range of the spatial positions of the new candidate individuals, thereby obtaining an updated mixed population; the partitioning module is used to partition the updated mixed population into a first population and a second population, and to update the first population to the antlion population and the second population to the ant population.

[0185] In some embodiments, the determining module is configured to determine the probability that the spatial location of the problematic individual is within a specific range of the spatial location of the new candidate individual; the determining module is further configured to determine the difference between the spatial location of the problematic individual and the spatial location of the new candidate individual; the updating module is configured to update the spatial location of the problematic individual based on the probability and the difference.

[0186] The descriptions of the above device embodiments are similar to those of the above method embodiments, and have similar beneficial effects. For technical details not disclosed in the device embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0187] It should be noted that, in the embodiments of this application... Figure 6 The hierarchical analysis device shown is illustrative of the module division, representing only one logical functional division; in actual implementation, other division methods may be used. Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, exist as separate physical units, or be integrated into one unit with two or more units. The integrated units can be implemented in hardware, as software functional units, or a combination of both.

[0188] It should be noted that, in the embodiments of this application, if the above-described methods are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to related technologies, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an electronic device to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks. Thus, the embodiments of this application are not limited to any specific hardware and software combination.

[0189] This application provides an electronic device. Figure 7 This is a schematic diagram of the hardware entity of the electronic device according to an embodiment of this application, such as... Figure 7 As shown, the electronic device 700 includes a memory 701 and a processor 702. The memory 701 stores a computer program that can run on the processor 702. When the processor 702 executes the program, it implements the steps in the method provided in the above embodiments.

[0190] It should be noted that the memory 701 is configured to store instructions and applications executable by the processor 702, and can also cache data to be processed or already processed by the processor 702 and various modules in the electronic device 700 (e.g., image data, audio data, voice communication data and video communication data), which can be implemented by flash memory or random access memory (RAM).

[0191] This application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method provided in the above embodiments.

[0192] This application provides a computer program product containing instructions that, when run on a computer, cause the computer to perform the steps in the method provided in the above-described method embodiments.

[0193] It should be noted that the descriptions of the storage medium and device embodiments above are similar to the descriptions of the method embodiments above, and have similar beneficial effects. For technical details not disclosed in the storage medium, storage medium, and device embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0194] It should be understood that the phrases "one embodiment," "an embodiment," or "some embodiments" mentioned throughout the specification mean that a specific feature, structure, or characteristic related to an embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment," "in one embodiment," or "in some embodiments" appearing throughout the specification do not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of this application, the sequence numbers of the above-described processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. The sequence numbers of the above-described embodiments are merely for descriptive purposes and do not represent the superiority or inferiority of the embodiments. The descriptions of the various embodiments above tend to emphasize the differences between the various embodiments; their similarities or commonalities can be referred to mutually, and for the sake of brevity, they will not be repeated here.

[0195] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that there can be three kinds of relationships. For example, object A and / or object B can represent three situations: object A exists alone, object A and object B exist simultaneously, and object B exists alone.

[0196] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0197] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple modules or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or modules can be electrical, mechanical, or other forms.

[0198] The modules described above as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules. They may be located in one place or distributed across multiple network units. Some or all of the modules may be selected to achieve the purpose of this embodiment according to actual needs.

[0199] In addition, each functional module in the various embodiments of this application can be integrated into one processing unit, or each module can be a separate unit, or two or more modules can be integrated into one unit; the integrated modules can be implemented in hardware or in the form of hardware plus software functional units.

[0200] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0201] Alternatively, if the integrated units described above are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to related technologies, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an electronic device to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0202] The methods disclosed in the several method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.

[0203] The features disclosed in the several product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0204] The features disclosed in the several method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method or device embodiments.

[0205] The above description is merely an embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A hierarchical analysis method, characterized in that, The method includes: Based on the attribute parameters of the current candidate solutions, establish a hierarchical structure model; Based on the hierarchical structure model, a judgment matrix is ​​constructed; wherein, the elements in the judgment matrix are used to characterize the relative importance among the attribute parameters; The judgment matrix is ​​subjected to antlion optimization to obtain a weight vector and a consistency index; wherein, the weight vector is used to characterize the relative importance of the current candidate solution in the candidate solution set, and the consistency index is used to characterize the rationality of the current candidate solution; The target solution is determined from the set of candidate solutions based on the weight vector and consistency index of each candidate solution. The step of performing antlion optimization on the judgment matrix to obtain the weight vector and consistency index includes: Based on the judgment matrix, an antlion population and an ant population are generated; wherein the population size of the antlion population and the population size of the ant population have the same dimension as the judgment matrix. The spatial location and fitness function value of the antlion population and each individual in the ant population are determined; wherein the fitness function value is used to characterize the degree of fitness of the individual in the population; Elite individuals are selected from the antlion population and the ant population based on the fitness function value of each individual. The spatial location of the elite individual is determined as the weight vector, and the fitness function value of the elite individual is determined as the consistency index.

2. The method according to claim 1, characterized in that, The step of selecting elite individuals from the antlion population and the ant population based on the fitness function value of each individual includes: Individuals in the antlion population whose fitness function values ​​meet the candidate criteria are identified as candidate individuals. Based on the spatial location of the candidate individuals, the ant population is updated to obtain the updated ant population. The updated ant population is mixed with the antlion population to obtain a mixed population; The first individual in the mixed population whose fitness function value satisfies the candidate condition is updated to obtain a new candidate individual; Return to the spatial location of the candidate individuals, update the ant population, obtain the updated ant population, and repeat the process until the number of updates meets the first convergence condition. Then, determine the new candidate individuals obtained based on the current mixed population as the elite individuals.

3. The method according to claim 2, characterized in that, The step of updating the ant population based on the spatial location of the candidate individuals to obtain the updated ant population includes: Update the spatial position of each individual in the ant colony based on the spatial position of the candidate individuals; The fitness function value of the corresponding individual in the ant colony is updated based on the updated spatial location of the individual in the ant colony.

4. The method according to claim 2, characterized in that, The step of updating the first individual in the mixed population whose fitness function value satisfies the candidate condition to obtain a new candidate individual includes: The spatial position of the first individual is subjected to chaotic optimization processing to obtain the spatial position of the second individual; The fitness function value of the second individual is determined based on its spatial location. Based on the fitness function value of the second individual and the fitness function value of the first individual, determine whether the second individual meets the update condition; If the second individual meets the update condition, the second individual is updated to the first individual; If the second individual does not meet the update conditions, the first individual is retained; Return to the spatial position of the first individual and perform chaotic optimization to obtain the spatial position of the second individual. Repeat this process until the number of optimizations meets the second convergence condition. Then, determine the first individual obtained so far as the new candidate individual.

5. The method according to claim 4, characterized in that, The process of performing chaotic optimization on the spatial position of the first individual to obtain the spatial position of the second individual includes: Map the spatial position of the first individual from the first space to the second space to obtain the mapped position of the second individual; The mapping position of the second individual is reverse mapped from the second space to the first space to obtain the spatial position of the second individual; wherein the spatial position of the second individual is different from the spatial position of the first individual.

6. The method according to claim 2, characterized in that, Before continuing to update the updated ant colony based on the spatial locations of the new candidate individuals, the method further includes: Identify undesirable individuals in the mixed population; The spatial locations of undesirable individuals in the mixed population are modified so that their spatial locations fall within a specific range of the spatial locations of the new candidate individuals, thereby obtaining an updated mixed population. The updated mixed population is divided into a first population and a second population, and the first population is updated to the antlion population, and the second population is updated to the ant population.

7. The method according to claim 6, characterized in that, The process of altering the spatial location of undesirable individuals in the mixed population includes: Determine the probability that the spatial location of the defective individual lies within a specific range of the spatial location of the new candidate individual; Determine the difference between the spatial location of the defective individual and the spatial location of the new candidate individual; The spatial location of the defective individual is updated based on the probability and the difference.

8. A hierarchical analysis device, characterized in that, include: The construction module is used to build a hierarchical model based on the attribute parameters of the current candidate solutions; The construction module is further configured to construct a judgment matrix based on the hierarchical structure model; wherein the elements in the judgment matrix are used to characterize the relative importance between the attribute parameters; The processing module is used to perform antlion optimization on the judgment matrix to obtain a weight vector and a consistency index; wherein, the weight vector is used to characterize the relative importance of the current candidate solution in the candidate solution set, and the consistency index is used to characterize the rationality of the current candidate solution; The determination module is used to determine the target solution from the set of candidate solutions based on the weight vector and consistency index of each candidate solution; A generation module is used to generate antlion populations and ant populations based on the judgment matrix; wherein the population size of the antlion population and the population size of the ant population have the same dimension as the judgment matrix. The determining module is further configured to determine the spatial location and fitness function value of the antlion population and each individual in the ant population; wherein the fitness function value is used to characterize the degree of adaptation of the individual in the population; The selection module is used to select elite individuals from the antlion population and the ant population based on the fitness function value of each individual. The determining module is further configured to determine the spatial location of the elite individual as the weight vector, and to determine the fitness function value of the elite individual as the consistency index.

9. An electronic device comprising a memory and a processor, the memory storing a computer program executable on the processor, characterized in that, When the processor executes the program, it implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 7.

Images