CN117349532B - Dynamic multi-target service combination optimization recommendation method and system - Google Patents
Dynamic multi-target service combination optimization recommendation method and system Download PDFInfo
- Publication number
- CN117349532B CN117349532B CN202311388437.2A CN202311388437A CN117349532B CN 117349532 B CN117349532 B CN 117349532B CN 202311388437 A CN202311388437 A CN 202311388437A CN 117349532 B CN117349532 B CN 117349532B
- Authority
- CN
- China
- Prior art keywords
- service
- population
- optimization
- environment
- combination
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 134
- 238000000034 method Methods 0.000 title claims abstract description 59
- 230000008859 change Effects 0.000 claims abstract description 128
- 230000004044 response Effects 0.000 claims abstract description 54
- 238000001514 detection method Methods 0.000 claims abstract description 32
- 230000009326 social learning Effects 0.000 claims abstract description 23
- 230000013016 learning Effects 0.000 claims description 52
- 230000007613 environmental effect Effects 0.000 claims description 50
- 239000000203 mixture Substances 0.000 claims description 21
- 230000006870 function Effects 0.000 claims description 18
- 230000035772 mutation Effects 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 14
- 238000013461 design Methods 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 2
- 238000002474 experimental method Methods 0.000 description 10
- 230000007246 mechanism Effects 0.000 description 7
- 108090000623 proteins and genes Proteins 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 4
- 230000007423 decrease Effects 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 230000013011 mating Effects 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000000717 retained effect Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000009849 deactivation Effects 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 229940060587 alpha e Drugs 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- DBOHWMPKJCJANT-UHFFFAOYSA-N brooker's merocyanine Chemical compound C1=CN(C)C=CC1=CC=C1C=CC(=O)C=C1 DBOHWMPKJCJANT-UHFFFAOYSA-N 0.000 description 1
- 210000000349 chromosome Anatomy 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000003340 mental effect Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000011895 specific detection Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- Entrepreneurship & Innovation (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Educational Administration (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of service recommendation, and provides a dynamic multi-objective service combination optimization recommendation method and system, wherein the method comprises the following steps: aiming at the dynamic multi-objective service optimization combination problem, constructing a dynamic multi-objective service optimization combination problem model; designing an environment change detection operator and an environment change response strategy aiming at the dynamic property of the environment; and embedding the environment change detection operator and the environment change response strategy into a social learning optimization algorithm, and solving a dynamic multi-objective service optimization combination problem model based on the embedded improved social learning optimization algorithm to obtain an optimal service combination recommendation scheme. The method and the system can sense the change of the environment in real time, and obtain the optimal and most effective service combination recommendation scheme matched with the corresponding environment.
Description
Technical Field
The disclosure relates to the technical field of service recommendation, in particular to a dynamic multi-objective service combination optimization recommendation method and system.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
Service composition is a new mode for creating value added services in a dynamic network environment, and the service composition comprises services with single functions into large granularity so as to meet the complex demands of users. The service portfolio optimization problem typically has multiple conflicting optimization objectives and in reality there are a large number of services that are functionally identical or similar but of different quality of service (Quality of Service, qoS). On the premise of meeting the functional and non-functional requirements of users, the efficient selection of the optimal combined service from the massive service combination schemes is an important research problem in the field of service computing.
In recent years, problems of dynamic changes in service combination environments, such as logistics service combination, travel service combination, senior citizen service combination, and the like, have emerged in real applications. Taking the insurance service combination problem as an example, the insurance service combination constructs a composite service by combining a plurality of individual insurance services to satisfy the complex insurance needs of the insurance user. In one aspect, the user's goals have dynamics, and the user's insuring goals may change at any stage, and the variability of the user's goals may make the built insurance service composition scheme fall short of the user's requirements. On the other hand, the insurance service resources have dynamics, including the addition of new insurance services, the change or the deactivation of old insurance services, etc., and a certain insurance service in the insurance service combination may not complete the corresponding combination task due to the abnormality. The candidate service resources of the service combination optimization problem can change, be newly added, be deactivated and the like along with the time or environment change, and simultaneously the user target can also change, so that the dynamic changes of an objective function, a solution space, constraint conditions and the like of the service combination optimization problem are caused, and the service combination optimization problem belongs to the dynamic multi-target optimization problem.
Currently available service composition research works can be summarized as a single-objective service composition optimization method and a multi-objective service composition optimization method. The single-objective service combination optimization method mainly aggregates a plurality of optimization objectives into a single-objective optimization problem through linear weighting and other methods. The multi-objective service combination optimization method mainly refers to an intelligent optimization method. The inventor finds in the research that the existing method ignores the dynamic property of the service combination and can not effectively solve the dynamic multi-objective service combination optimization problem. When service resources are newly added, changed and deactivated, or the original target of the user is changed, service combination cannot be performed again according to different changes, so that the final service combination scheme is unavailable, non-optimal or does not meet the target requirement of the user.
Disclosure of Invention
In order to solve the above problems, the disclosure provides a dynamic multi-objective service combination optimization recommendation method and system, which can sense the change of the environment in real time and obtain an optimal and most effective service combination recommendation scheme matched with the corresponding environment.
In order to achieve the above purpose, the present disclosure adopts the following technical scheme:
One or more embodiments provide a dynamic multi-objective service combination optimization recommendation method, including the following steps:
Aiming at the dynamic multi-objective service optimization combination problem, constructing a dynamic multi-objective service optimization combination problem model;
designing an environment change detection operator and an environment change response strategy aiming at the dynamic property of the environment;
and embedding the environment change detection operator and the environment change response strategy into a social learning optimization algorithm, and solving a dynamic multi-objective service optimization combination problem model based on the embedded improved social learning optimization algorithm to obtain an optimal service combination recommendation scheme.
One or more embodiments provide a dynamic multi-objective service composition optimization recommendation system, comprising:
Model construction module: the method comprises the steps of being configured to construct a dynamic multi-objective service optimization combination problem model aiming at the dynamic multi-objective service optimization combination problem;
and (3) environment detection and corresponding design modules: configured to design an environmental change detection operator and an environmental change response policy for the dynamics of the environment;
And a solving module: the method comprises the steps of embedding an environment change detection operator and an environment change response strategy into a social learning optimization algorithm, and solving a dynamic multi-objective service optimization combination problem model based on the embedded improved social learning optimization algorithm to obtain an optimal service combination recommendation scheme.
An electronic device comprising a memory and a processor and computer instructions stored on the memory and running on the processor, which when executed by the processor, perform the steps of a dynamic multi-objective service composition optimization recommendation method as described above.
A computer readable storage medium storing computer instructions which, when executed by a processor, perform the steps of a dynamic multi-objective service composition optimization recommendation method as described above.
Compared with the prior art, the beneficial effects of the present disclosure are:
In the method, a dynamic multi-objective service combination problem model is established, a social learning optimization algorithm (Social Learning Optimization, SLO) is improved, a new learning operator is provided, an environment change detection operator is designed, an environment change response strategy combining a prediction mechanism and a diversity introduction mechanism is provided, the environment change detection operator and the environment change response strategy are embedded into the improved social learning optimization algorithm, when the environment changes, changing environment factors can be embedded into a solving process, an optimal service combination corresponding to the changed new environment can be obtained, the effectiveness of service combination recommendation is improved, the usability of a combination scheme can be ensured, the target requirements of users are met, and the method is an optimal service combination recommendation scheme.
The advantages of the present disclosure, as well as those of additional aspects, will be described in detail in the following detailed description of embodiments.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure, illustrate and explain the exemplary embodiments of the disclosure and together with the description serve to explain and do not limit the disclosure.
FIG. 1 is a flow chart of a dynamic multi-objective service composition optimization recommendation method of embodiment 1 of the present disclosure;
FIG. 2 is a service composition workflow of embodiment 1 of the present disclosure;
FIG. 3 is a service encoding scheme of embodiment 1 of the present disclosure;
FIG. 4 is a graph showing the comparison of experimental results of different iterations of the DMOSCO algorithm of example 1 with different numbers of candidate services;
FIG. 5 is a graph showing the comparison of the experimental results of different iteration times of the same candidate service number under different subtasks using DMOSCO algorithm of the present embodiment 1;
fig. 6 is the QoS values of service combination solutions searched for at different problem scales using the DMOSCO algorithm of this embodiment 1;
FIG. 7 shows QoS values for different environmental changes using DMOSCO algorithm of example 1;
FIG. 8 is a comparison of results of solving the same dynamic multi-objective service optimization combination problem using the prior art method and DMOSCO algorithm;
FIG. 9 is a graph comparing the result of solving the multi-objective service optimization combination problem by adopting DMOSCO algorithm and original SLO algorithm of the present embodiment;
FIG. 10 is a graph of comparison of the results of an algorithm solution using a different environmental response strategy DMOSCO.
Detailed Description
The disclosure is further described below with reference to the drawings and examples.
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present disclosure. Unless defined otherwise, 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 disclosure belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments in accordance with the present disclosure. As used herein, the singular is also intended to include the plural unless the context clearly indicates otherwise, and furthermore, it is to be understood that the terms "comprises" and/or "comprising" when used in this specification are taken to specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof. It should be noted that, without conflict, the various embodiments and features of the embodiments in the present disclosure may be combined with each other. The embodiments will be described in detail below with reference to the accompanying drawings.
Example 1
In one or more embodiments, as shown in fig. 1 to 10, a Dynamic Multi-objective service combination optimization recommendation method (Dynamic Multi-object Service Composition Optimization, DMOSCO) includes the following steps:
Step 1, constructing a dynamic multi-objective service combination optimization problem model aiming at the dynamic multi-objective service combination optimization problem;
Step 2, designing an environment change detection operator and an environment change response strategy aiming at the dynamic property of the environment;
Specifically, detecting an environmental change based on an environmental change detection operator; the environmental change response strategy predicts the moving direction of the population by using the moving step length of the representative individuals according to the environmental change, and carries out diversity adjustment on the predicted new population;
And step 3, embedding an environment change detection operator and an environment change response strategy into a social learning optimization algorithm, and solving a dynamic multi-objective service combination optimization problem model based on the improved social learning optimization algorithm to obtain an optimal service combination recommendation scheme.
In the embodiment, a social learning optimization algorithm (Social Learning Optimization, SLO) is improved, a new learning operator is provided, an environment change detection operator and an environment change response strategy are embedded into the improved social learning optimization algorithm, when the environment changes, the changed environment factors can be embedded into a solving process, an optimal service combination corresponding to the changed new environment can be obtained, the effectiveness of service combination recommendation is improved, the availability of a combination scheme can be ensured, the target requirements of a user are met, and the service combination recommendation scheme is optimal.
The Dynamic Multi-objective optimization problem (Dynamic Multi-objective Optimization Problems, DMOPs) refers to a Multi-objective optimization problem that an objective function and a constraint condition change with time, and service combination optimization problems in some fields have Dynamic characteristics, such as a user objective, a user preference, a service resource and the like may change with time, so that the objective function and the constraint condition of the optimization problem change with time. Therefore, in this embodiment, it is proposed to construct a dynamic multi-objective optimization model to solve such service combination optimization problem.
In step 1, service combination refers to decomposing a task into abstract subtasks with fine granularity and single function according to the multifunctional service requirements submitted by users, selecting candidate services corresponding to each subtask from a candidate service set meeting the requirements, and quickly combining the candidate services to form complete combined services so as to meet different user requirements.
Specifically, the service composition requirement wsc= { T 1,T2,...,Tn } is composed of n subtasks, where T i represents the ith subtask after the task is split. CS 1,CS2,...,CSn represents a candidate service set for subtasks in a workflow, whereRepresenting a set of m candidate services corresponding to each subtask T i. For each candidate service/>The candidate service corresponds to r QoS attribute values, denoted qos= { QoS 1,qos2,...,qosr }. The service composition workflow is shown in fig. 2.
In the service composition problem based on the workflow method, the execution relationship of the subtasks may include: four types of sequential, parallel, selection, and rotation. As shown in Table 1, there is provided a QoS calculation method of 4 flow structures composed of n service nodes, where n is the number of sub-tasks and p i represents the probability that the ith sub-service set is selected and satisfies
Table 1 combined service QoS calculating method
In table 1, cost i represents a QoS attribute value of a Cost of a sub-service, RT i represents a QoS attribute value of a response time, RE i represents a QoS attribute value of reliability, and l represents the number of loops.
In the service combination optimization process, service resources may be newly added, changed or deactivated, and the requirements of users may also be changed. On one hand, the cost, response time and reliability function value of the model can be dynamically changed along with the new increase, change or deactivation of the service resource, so that the optimization objective function of the model is changed; on the other hand, the user's needs may also change, resulting in changes in the constraints and solution space of the model.
In this embodiment, the service optimization combination problem is modeled as a dynamic multi-objective service optimization combination model, and three QoS attributes which are opposite to each other and are important in service combination are selected as optimization objectives, including cost, response time and reliability.
In this embodiment, the dynamic multi-objective service optimization combination model has the objective function of lowest cost, shortest response time and highest reliability, and the global constraint condition is: the total Cost does not exceed the Cost maximum value Cost max, the response Time does not exceed the response Time maximum value Time max, the reliability is not lower than the set reliability threshold Rel min, and the mathematical model of the dynamic multi-objective service combination optimization problem is:
Wherein F (x, t) is a three-dimensional objective function, x= { x 1,x2,...,xn } is an n-dimensional decision vector in the decision space Ω;
t is a time variable, and in this embodiment, the dynamic characteristic of the objective function changing with time is described based on the constructed time variable, which is defined as follows:
Where τ is the number of iterations, τ t is the frequency of the environmental change, n t is the strength of the environmental change, which includes the service resource change and the user demand change.
Optionally, the environmental change frequency is set to a fixed value by referring to other experiments;
Optionally, the environmental change intensity sets different levels of change intensity according to different changes of service resources and user demands.
In this embodiment, the Pareto dominant concept is used to calculate the fitness value of the individual to evaluate the solution, if Individual1 is greater than Individual2, individual is considered to be better than Individual, and the dominant solution number dom (i) of the solution in the population is increased by 1, where the fitness value of the solution in this embodiment is calculated as follows:
Where dom (i) is the number of dominant solutions and PS is the number of solutions in the Pareto optimal solution set.
The embodiment provides a dynamic multi-target service combination optimization problem and establishes a dynamic multi-target service combination optimization problem model. Under a dynamic environment, the optimization target, the value space and the constraint condition of the service combination optimization problem can dynamically change along with time and environmental changes, the service combination optimization problem constructed by the embodiment is more in line with the actual situation, the environmental changes can be timely detected, the service combination optimization problem can dynamically change along with the environmental changes, the accuracy of a problem model is improved, and therefore a more reasonable service combination optimization scheme can be provided.
In step 2, the environmental change detection includes: user target change and service resource change;
Aiming at the change of the user target, whether the user target is changed or not is judged by detecting whether the user target at the moment t-1 at the last moment is the same as the user target at the current moment t;
For service resource change, whether the service resource is changed is judged by detecting whether the candidate service at the current time t is available, whether the candidate service attribute values of the previous time t-1 and the current time t are the same and whether there is a newly added candidate service resource.
Specifically, the specific detection mode of the environmental change detection is as follows:
if the user target is detected to be changed, changing constraint conditions; and if the service resource is detected to be changed, executing a response strategy.
When detecting that the service resource changes, the environment change response strategy comprises the following steps:
Step 21, selecting a central point of the current moment t and an extreme point on each objective function to form a representative individual set;
wherein PS is the Pareto optimal solution set.
For the minimum problem, a boundary point is an individual with a minimum in a certain dimension in the target space. The number of boundary points is equal to the dimension of the target space, and the calculation formula of the PS center point at the moment t is as follows:
wherein, Is the ith dimension variable of the PS center point at time t, PS t is the Pareto solution set at time t, |ps t | is the number of solutions in PS t,/>Is the i-th dimension variable of a solution in PS t.
Step 22, calculating the distance from all the individuals in the population to each representative individual, and associating the individual with the representative individual with the smallest distance;
x i＝(xi1,xi2,...,xin) is any one of the population, To represent any one of a set of individuals, X i and/>The distance calculation formula of (2) is as follows:
Step 23, predicting a population at the next time t+1 according to the evolution step length of the individual represented at the current time t;
specifically, when the environment changes, the evolution direction of the related individuals is predicted by recording the front-back position change of each representative individual, and then And/>The evolution step/>, is a representative set of individuals at time t and time t-1 respectivelyThe following can be defined:
where i=1, 2, e, e represents the number of representative individuals by representative individuals The evolutionary step length of the model is calculated, the position of other individuals associated with the model at the time t+1 realizes the prediction of the population at the time t+1, and the calculation formula is as follows:
And step 24, randomly selecting individuals with set proportions in the predicted population to execute a diversity introduction strategy, and taking the obtained predicted population as an initial population in a new environment.
In order to increase the diversity of the population, avoiding the population from being trapped in a local optimum due to the guidance of the representative individuals, the diversity introduction strategy in this embodiment is: initializing individuals with set proportions in the prediction population, wherein the specific operation is as follows:
Wherein X i＝(xi1,xi2,...,xin) is any one of the population, M represents the number of candidate service numbers corresponding to the ith subtask; gamma is the probability of selection.
Aiming at the problem of dynamic environment change, the embodiment provides an environment change response strategy. The strategy combines a prediction mechanism with a diversity introduction mechanism, predicts the moving direction of a population by using the moving step length of a representative individual, and carries out diversity adjustment on a predicted new population, and both the population convergence and the diversity are considered, so that an algorithm can timely detect the change of the environment and make effective response.
The social learning optimization algorithm (Social Learning Optimization Algorithm, SLO) is a novel group intelligent algorithm simulating the human social intelligent evolution process. SLO is composed of co-evolution micro-space, learning space and belief space, and has a better optimization mechanism.
The existing SLO cannot be directly used for solving the dynamic multi-objective service combination optimization problem, and an operator suitable for the dynamic multi-objective optimization problem is designed for the embodiment;
Step 3, embedding an environment change detection operator and an environment change response strategy into an improved social learning optimization algorithm to form a solution method for a dynamic multi-objective service combination optimization problem, wherein the overall flow of the method is shown in fig. 4, and the method comprises the following steps:
step 31, performing service combination coding, namely coding a feasible service combination scheme into an individual in the SLO, and constructing an initial population;
Specifically, N initial points are randomly generated in a decision space to form an initial population P 0, and a population scale N, a maximum iteration number T maxgen, a mutation rate P m, an intersection rate P c, the number of subtasks N and a candidate service number m are set;
and 32, executing an environment change detection process, and when an environment change execution environment change response strategy is identified, forming a prediction population under a new environment, wherein the generation of the prediction population is as follows:
Pt＝changeStrategy(Pt-1)
Step 33, sequentially executing selection operation, crossover operation and mutation operation in the micro-space aiming at the obtained predicted population, and selecting better individuals to form the population obtained by operation in the micro-space;
the crossover operation is as follows: p t mating＝Mating(Pt)
The mutation operation is as follows: p t mutation＝Mutation(Pt mating)
Step 34, aiming at the population obtained by operation in the micro-space, observation learning and imitation learning in the learning space are sequentially executed, and better individuals are selected to form the optimized population in the learning space;
Observation study: p t observation＝Observational(Pt mutation)
Imitation learning: p t+1＝Imitation(Pt observation)
Step 35, aiming at the population obtained by optimization in the learning space, updating operation and influencing operation in the belief space are executed, and the population obtained by optimization in the belief space is obtained;
Step 36, circularly executing the steps 32 to 35 until the set iteration times are reached, outputting an optimal population, wherein individuals in the population are the optimal solution sets of the service combinations.
The following describes the optimization operations in each evolutionary space in the improved SLO, respectively.
(1) Service combination coding: an individual in the SLO represents a viable service composition scheme;
Population individuals represented by n-dimensional vectors in SLO are viable solutions to the optimization problem, with individual quality represented by non-dominant ordering relationships of the relevant solutions. When used to solve a service portfolio optimization problem, one possible service portfolio scenario can be represented by an individual in the SLO, each dimension of which must be an integer that satisfies the boundary condition.
Specifically, the individual codes adopt integer coding modes, and the specific coding modes are shown in fig. 3, whereinThe m candidate services corresponding to the nth subtask are represented, and the genes in the codes represent discrete service serial numbers.
Assuming that the population size of the algorithm is N, the population is p= { X 1,...,Xk,...,XN }, the individual X k is an N-dimensional array X k＝(xk1,...,xki,...,xkn), where the element X ki is a service sequence number for executing the ith subtask in the kth service combination scheme, X ki e {1,2, 3..m }, and m represents the number of candidate services corresponding to the ith subtask.
(2) Operations within the subspace: including selection, crossover and mutation operations;
Alternatively, a roulette wheel selection method is used as a selection operation;
Alternatively, the interleaving operation is: randomly selecting positions to perform cross operation, and executing greedy operation to select the optimal individual;
specifically, let X i＝(xi1,...,xik,...,xin) and X j＝(xj1,...,xjk,...,xjn) be any two different individuals, r be a random number in the interval (0, 1), p c be the crossing rate, q be a random integer in the interval [1, n ], and represent the crossing position of the two individuals. If r < p c, then a cross operation is performed. After the crossover operation, the crossover operation is performed on individuals X i and Between, X j and/>And performing greedy operation and reserving better individuals, wherein the cross operation is shown in a formula (10).
Alternatively, the interleaving operation is: randomly selecting a gene point of an individual, randomly generating a new gene within a set range, replacing the randomly selected gene point, and retaining the individual with higher fitness value.
Specifically, let X i＝(xi1,...,xik,...,xin) be any one of the population, p m be the mutation rate, randomly selecting a mutation gene point X ik in the chromosome of the current individual, and randomly generating a new gene X ih within the range of {1,2,3,..m } to replace the mutation gene point X ik in X i. After the mutation operation, the individual with higher fitness value is retained by greedy operation, and the mutation operation is shown in formula (11).
(3) Operations within the learning space: including observation learning and imitation learning;
operations within the learning space include observation learning and imitation learning. Observational learning refers to an individual improving mental capacity by observing the behavior of excellent individuals and learning their advantages. Imitation learning refers to an individual randomly imitating surrounding individuals.
The learning operator in the improved SLO is defined as follows:
a. Observation study
When an individual performs observation and learning, a part of own knowledge is reserved due to inertia. In this embodiment, the mechanism is implemented by introducing an inertia coefficient and a disturbance learning factor based on a Sigmoid function in the observation learning operation.
Specifically, X i＝(xi1,...,xik,...,xin) is set as any one of the population,For the best individual,/>To observe a new individual generated after learning, the observation learning operation is defined as follows:
Wherein ω is learning inertia weight, ω∈ (0, 1), ω·x ik represents self-retaining portion of the individual in observation learning; (1/1+e -α) is a learning perturbation factor based on Sigmoid function, alpha E < -6,6 >, Representing the part obtained after the individual learns to the optimal individual; [] Is a rounding operation. After observing the learning operation, individuals with higher fitness values are retained using a greedy operation.
B. imitation learning: combining a non-dominant individual with two random individuals to form a simulated learning group for simulated learning;
specifically, X i＝(xi1,...,xik,...,xin) is set as any body in the population to carry out imitation learning operation, And/>In the case of a random individual,For any one of the non-dominant solutions,/>To simulate a new individual generated after a learning operation, the definition of simulated learning is as follows:
Wherein r 1,r2 e {1,., N }, N is population size, and r 1≠r2; s.epsilon.0, 1 is the scaling factor, and [ (O) is the rounding operation. After mimicking the learning operation, individuals with higher fitness values are retained using a greedy operation.
(4) Operations within the belief space include knowledge updating operations and cultural affecting operations;
belief space simulates the process that human intelligent evolution is influenced by culture, including knowledge updating operation and culture influencing operation. The knowledge updating operation and the cultural affecting operation are defined as follows:
a. knowledge updating operation:
the knowledge updating operation refers to updating and accumulating knowledge for individuals in the belief space when the learning space obtains new excellent individuals. The update operation formula is as follows:
α＝N*β (14)
Wherein a refers to the number of excellent individuals in the current population to be selected; n represents population size; beta is the probability of selection.
B. Cultural impact operation:
the cultural influence operation refers to using knowledge in belief space to influence individuals with low non-dominant ranking level in micro space, so as to guide the group to evolve to a better direction and improve the convergence speed of the algorithm. The influencing operation formula is as follows:
wherein, X i＝(xi1,...,xik,...,xin) is any one of the population; a j is any individual in belief space; t is the current iteration number; epsilon is the update interval parameter.
Further, the method further comprises a correction operation step of correcting the new individual after the learning operation in the learning space so that the number of candidate services of the individual does not exceed the total number of candidate services.
The value of each dimension variable of all individuals is a candidate service serial number corresponding to the subtask, and in a new individual generated by observation learning and imitation learning, the value of a dimension variable may exceed the value range, and the dimension variable is corrected according to the following formula, wherein m is the total number of candidate services corresponding to the subtask.
Aiming at the characteristics of the dynamic multi-objective service combination optimization problem, the embodiment designs an operator suitable for the dynamic multi-objective service combination optimization problem, and improves the solving quality of the service combination optimization problem. The proposed environment change detection operator and response strategy are combined to form a solution method for the dynamic multi-objective service combination optimization problem.
To illustrate the effect of the method of this example, experiments and analyses were performed.
The four groups of experiments are designed for respectively analyzing DMOSCO the feasibility and adaptability of solving the dynamic multi-objective service combination optimization problem, comparing DMOSCO the performances of other two algorithms in a dynamic environment, analyzing the effectiveness of an improvement strategy of an SLO algorithm and analyzing the effectiveness of an environment change response strategy.
1. Experimental conditions
Dynamic is a basic feature of a service composition environment, and its dynamic types can be divided into service resource changes and user target changes. When the environment change is detected, an environment change response mechanism needs to be executed to adjust the population, and the greater the environment change degree is, the greater the adjustment difficulty is. The experiment changes the attribute value of certain atomic services or sets the attribute value of the atomic services to be unavailable to simulate the change of service resources through randomly selecting certain atomic services in the iteration process, and simulates the change of a user target through randomly modifying constraint conditions or objective functions in the iteration process. In combination of these two cases, the environmental change degree was set to four kinds of grades of 0%, 1%, 5% and 10%, and the correspondence between the environmental change grade and the environmental change is shown in table 2.
TABLE 2 correspondence between environmental change conditions and environmental change levels
| Varying the grade | Service resource variation ratio | Number of constraint condition changes | Number of user target changes |
| 0％ | 0 | 0 | 0 |
| 1％ | 1％ | 1 | 0 |
| 5％ | 5％ | 2 | 1 |
| 10％ | 10％ | 3 | 2 |
The experiment adopts a random data set, each candidate service is set to have three QoS attributes of cost, response time and reliability, the QoS attribute values are randomly generated within a certain range, wherein the cost of the service is between 10 and 100, the response time of the service is between 10 and 50, and the reliability of the service is between 0.5 and 1. Since the data set is generated in a random manner, each group of experiments is run 20 times respectively, and the average value of the 20 test results is selected as the final experiment result in order to eliminate the randomness of the final experiment result.
The experiment uses Java programming language to realize the dynamic multi-objective service combination optimization method based on improved SLO, wherein the crossover probability is set to 0.85, the mutation probability is set to 0.05, the knowledge update selection proportion is set to 10%, and the culture influence update interval is set to 5. The experimental environment was set up as PC,11th Gen Intel (R) Core (TM) i5-1135G7@2.40GHz 2.42GHz,Windows 10 (64 bits).
2. Experimental results and analysis
1. DMOSCO solve the performance of dynamic multi-objective service portfolio optimization problem:
(1) Verifying DMOSCO performance at different iteration times
In order to verify the feasibility of DMOSCO algorithm to solve the dynamic multi-objective service combination optimization problem under different iteration times of different candidate service numbers and different subtasks, the subtask number is set to be fixed to 10, the candidate service numbers corresponding to the subtasks are respectively set to 10, 20, 50, 100 and 200, and the running time of the algorithm under different iteration times is recorded. The experimental result is shown in fig. 4, the abscissa represents the iteration number of the algorithm, the ordinate represents the running time of the algorithm, and m represents the number of candidate services corresponding to each subtask.
As can be seen from fig. 4, when the number of subtasks is 10, the running time of DMOSCO algorithm increases with the number of iterations, while there are conditions of up-and-down fluctuations. In addition, the runtime of DMOSCO algorithm has no obvious correlation with different number of candidate services, and the runtime of DMOSCO algorithm does not increase exponentially as the number of candidate services increases.
Setting the number of candidate services corresponding to the subtasks to be fixed to 50, setting the number of the subtasks to be 5, 10, 15, 20 and 25 respectively, and recording the running time of the algorithm under different iteration times. The experimental result is shown in fig. 5, the abscissa represents the iteration number of the algorithm, the ordinate represents the running time of the algorithm, and n represents the number of subtasks.
As can be seen from fig. 5, when the number of candidate services corresponding to the subtasks is 50, the running time of DMOSCO algorithm increases with the number of iterations. When the iteration number is 100-500, the running time of DMOSCO algorithm and different subtasks do not show obvious correlation. When the iteration number is 500-800, the running time of DMOSCO algorithm and the number of different subtasks are in linear correlation. Furthermore, as the number of subtasks increases, the run time of the algorithm does not increase exponentially. In summary, it can be derived DMOSCO that it is feasible for the algorithm to solve the dynamic multi-objective service optimization combination problem.
(2) Verifying DMOSCO performance at different problem scales
In order to verify the adaptability of DMOSCO algorithm to solve the dynamic multi-objective service combination optimization problem under different problem scales, different problem scales are set by setting the number of different subtasks and the number of candidate services corresponding to the different subtasks. The experiment sets up five question scales, the number of subtasks and the number of candidate services are 5×10, 10×20, 20×50, 30×100, 50×200, respectively. The iteration number is set to 1000, and QoS values of service combination solutions searched by the algorithm under different problem scales are recorded. As shown in fig. 6, the experimental results are shown in the abscissa representing the problem scale, the ordinate representing the QoS value of the combined service solution set, and cost, time, reliability representing the cost, response time, and reliability of the combined service solution set, respectively.
As can be seen from fig. 6, the cost and time of the service portfolio solution searched DMOSCO decreases as the problem size increases, because as the problem size increases, the number of constituent portfolio services also increases, providing more and better choices for the algorithm. The reliability of the solution set increases continuously as the problem size increases from 5 x 10 to 20 x 50, and decreases continuously as the problem size increases from 20 x 50 to 50 x 200, indicating that the reliability of the solution set increases with the problem size and has a certain volatility. The cost and response time of service portfolio solutions do not decrease exponentially with increasing problem size, even with increasing reliability. Therefore, DMOSCO algorithm has certain adaptability to solve the problem of dynamic multi-objective service optimization combination.
(3) Verify DMOSCO performance under different environmental changes
In order to verify the performance of the method according to the present embodiment in response to different environmental change degrees, the iteration number is set to 1000, the number of subtasks is set to 10, the number of candidate services of the subtasks is set to 50, and the QoS values of the service combination solutions searched by the algorithm under the conditions that the environmental change degrees are 0%, 1%, 5% and 10% respectively are recorded. As shown in fig. 7, the experimental results are shown in the horizontal axis indicating the environmental change degree, the vertical axis indicating the QoS value of the combined service solution set, and cost, time, reliability indicating the cost, response time, and reliability of the combined service solution set, respectively.
Figure 7 shows the robustness and adaptability of DMOSCO to random changes in the environment. As the intensity of the environment change increases, the cost and time attribute values of the service combination solution set do not increase exponentially. Meanwhile, when the environmental change degree is 1%, the cost and time attribute values of the combined service solution set become better. Furthermore, the reliability attribute value of the service composition solution set is not exponentially reduced. Therefore, DMOSCO can quickly respond to the change of the environment when the environment is changed, and a new service combination scheme can be robustly generated without manually adjusting parameters.
2. DMOSCO algorithm performance comparison:
To verify the performance of DMOSCO algorithm, the same dynamic multi-objective service optimization combination problem is solved using DNSGA-II-A [32], MOED/D-FPS [33] and DMOSCO algorithms. The number of subtasks of the four algorithms is set to 10, the number of candidate services corresponding to the subtasks is set to 50, the initial population size is set to 100, the environmental change degree is set to 1%, and the adaptability values of service combination scheme solutions searched by the three algorithms under different iteration times are recorded. The experimental result is shown in fig. 8, and the abscissa represents the iteration number of the algorithm and the ordinate represents the fitness value of the solution set.
As can be seen from fig. 8, under the same environment, the DMOSCO algorithm converges faster than other algorithms when the environment changes, while some solutions searched by the DMOSCO algorithm are superior to the other two algorithms. Based on the experimental results, a conclusion can be drawn that the DMOSCO algorithm can efficiently solve the dynamic multi-objective service optimization combination problem, and has strong competitiveness compared with other algorithms.
3. Validating effectiveness of improved SLO policies
To verify the validity of the improved SLO strategy, the original SLO initialization mode, objective function and constraints are modified to solve the multi-objective service optimization combination problem, and SLO is compared with DMOSCO. The number of subtasks of the two algorithms is set to 10, the number of candidate services corresponding to the subtasks is set to 50, the initial population size is set to 100, the environmental change degree is set to 1%, and the adaptability values of service combination scheme solutions searched by the two algorithms under different iteration times are recorded. The experimental result is shown in fig. 9, and the abscissa represents the iteration number of the algorithm and the ordinate represents the fitness value of the solution set.
As can be seen from fig. 9, under the same dynamic service combination environment, jie Yuanyuan obtained by the DMOSCO algorithm is better than SLO, and the convergence rate of the DMOSCO algorithm is faster than that of the SLO algorithm, so that the improvement strategy proposed for SLO can be proved to be effective.
4. Verifying validity of environmental change response policies
To verify the effectiveness of the proposed environmental change response strategy, a predictive population is formed in the new environment using a random initialization strategy instead of the proposed environmental change response strategy in DMOSCO, i.e., when the environment changes, an environmental response strategy of 20% of individuals is randomly initialized, a variant called DMOSCO-PS, and DMOCSO is compared to DMOSCO-PS. The number of subtasks of the two algorithms is set to 10, the number of candidate services corresponding to the subtasks is set to 50, the initial population size is set to 100, the iteration number is set to 1000, and the adaptability values of service combination solutions searched by the two algorithms under different environmental change degrees are recorded. The experimental result is shown in fig. 10, the abscissa represents the environmental change degree of the algorithm, and the ordinate represents the fitness value of the solution sets of the two algorithms.
As can be seen from fig. 10, the fitness value of the DMOSCO searched solution is better than DMOSCO-PS. As the degree of environmental change increases, the gap between the DMOSCO-PS and DMOSCO algorithms searching for fitness values of the solution sets increases. Based on the experimental results, the validity of the provided environment change response strategy can be verified.
Experimental results show that DMOSCO for solving the dynamic multi-objective service combination optimization problem has better performance. DMOSCO can be applied to more complex service portfolio optimization problems, as well as incorporating decision maker preferences into the dynamic search process to achieve preference-aware dynamic multi-objective service portfolio optimization.
Example 2
Based on embodiment 1, in this embodiment, a dynamic multi-objective service combination optimization recommendation system is provided, including:
Model construction module: the method comprises the steps of being configured to construct a dynamic multi-objective service optimization combination problem model aiming at the dynamic multi-objective service optimization combination problem;
and (3) environment detection and corresponding design modules: configured to design an environmental change detection operator and an environmental change response policy for the dynamics of the environment;
And a solving module: the method comprises the steps of embedding an environment change detection operator and an environment change response strategy into a social learning optimization algorithm, and solving a dynamic multi-objective service optimization combination problem model based on the embedded improved social learning optimization algorithm to obtain an optimal service combination recommendation scheme.
Here, the modules in this embodiment are in one-to-one correspondence with the steps in embodiment 1, and the implementation process is the same, which is not described here.
Example 3
The present embodiment provides an electronic device including a memory and a processor, and computer instructions stored on the memory and running on the processor, which when executed by the processor, perform the steps in a dynamic multi-objective service combination optimization recommendation method of embodiment 1.
Example 4
The present embodiment provides a computer readable storage medium storing computer instructions that, when executed by a processor, perform steps in a dynamic multi-objective service combination optimization recommendation method of embodiment 1.
The foregoing description of the preferred embodiments of the present disclosure is provided only and not intended to limit the disclosure so that various modifications and changes may be made to the present disclosure by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure.
While the specific embodiments of the present disclosure have been described above with reference to the drawings, it should be understood that the present disclosure is not limited to the embodiments, and that various modifications and changes can be made by one skilled in the art without inventive effort on the basis of the technical solutions of the present disclosure while remaining within the scope of the present disclosure.
Claims (6)
1. The dynamic multi-objective service combination optimization recommendation method is characterized by comprising the following steps of:
Aiming at the dynamic multi-objective service optimization combination problem, constructing a dynamic multi-objective service optimization combination problem model;
designing an environment change detection operator and an environment change response strategy aiming at the dynamic property of the environment;
embedding an environment change detection operator and an environment change response strategy into a social learning optimization algorithm, and solving a dynamic multi-objective service optimization combination problem model based on the embedded improved social learning optimization algorithm to obtain an optimal service combination recommendation scheme;
detecting an environmental change based on an environmental change detection operator; the environmental change response strategy predicts the moving direction of the population by using the moving step length of the representative individuals according to the environmental change, and carries out diversity adjustment on the predicted new population;
the environmental change detection includes: user target change and service resource change;
Aiming at the change of the user target, judging whether the user target changes or not by detecting whether the user target at the previous moment is the same as the user target at the current moment t;
For service resource change, whether the service resource is changed is judged by detecting whether the candidate service at the current time t is available, whether the attribute value of the candidate service at the previous time is the same as that of the candidate service at the current time, and whether newly added candidate service resource exists;
When detecting that the service resource changes, the environment change response strategy comprises the following steps:
selecting a central point of the current moment PS and forming a representative individual set by extreme points on each objective function;
Calculating the distance from all individuals in the population to each representative individual, and associating the individual with the representative individual with the smallest distance;
predicting the population at the next moment according to the evolution step length of the individual represented at the current moment;
Randomly selecting individuals with set proportions in the predicted population to execute a diversity introduction strategy, and taking the obtained predicted population as an initial population under a new environment;
the diversity introduction strategy is: initializing individuals with set proportions in the prediction population;
PS is Pareto optimal solution set;
the solving process of the optimal service combination recommendation scheme comprises the following steps:
Performing service combination coding, namely coding a feasible service combination scheme into an individual in SLO, and constructing an initial population;
executing an environment change detection process, and when the environment change is identified, executing an environment change response strategy to form a prediction population under a new environment;
sequentially executing selection operation, cross operation and mutation operation in the micro-space aiming at the obtained predicted population, and selecting more optimal individuals to form the population obtained by operation in the micro-space;
Aiming at the population obtained by operation in the micro space, observation learning and imitation learning in the learning space are sequentially executed, and better individuals are selected to form the population obtained by optimization in the learning space;
aiming at the population obtained by optimization in the learning space, updating operation and influencing operation in the belief space are executed, and the population after the belief space optimization is obtained;
and iteratively executing the steps until the set iteration times are reached, outputting an optimal population, and obtaining an optimal solution set of the service combination.
2. The method for optimizing and recommending dynamic multi-objective service combinations according to claim 1, wherein: the dynamic multi-objective service optimization combination model has the advantages that an objective function is the lowest cost, the shortest response time and the highest reliability, and the global constraint condition is as follows: the total cost does not exceed the cost maximum, the response time does not exceed the response time maximum, and the reliability is not lower than the set reliability threshold.
3. The dynamic multi-objective service combination optimization recommendation method according to claim 1, further comprising a correction operation step of correcting the new individual after the learning operation in the learning space so that the number of candidate services of the individual does not exceed the total number of candidate services.
4. A dynamic multi-objective service composition optimization recommendation system, comprising:
Model construction module: the method comprises the steps of being configured to construct a dynamic multi-objective service optimization combination problem model aiming at the dynamic multi-objective service optimization combination problem;
and (3) environment detection and corresponding design modules: configured to design an environmental change detection operator and an environmental change response policy for the dynamics of the environment;
and a solving module: the method comprises the steps of embedding an environment change detection operator and an environment change response strategy into a social learning optimization algorithm, and solving a dynamic multi-objective service optimization combination problem model based on the embedded improved social learning optimization algorithm to obtain an optimal service combination recommendation scheme;
detecting an environmental change based on an environmental change detection operator; the environmental change response strategy predicts the moving direction of the population by using the moving step length of the representative individuals according to the environmental change, and carries out diversity adjustment on the predicted new population;
the environmental change detection includes: user target change and service resource change;
Aiming at the change of the user target, judging whether the user target changes or not by detecting whether the user target at the previous moment is the same as the user target at the current moment t;
For service resource change, whether the service resource is changed is judged by detecting whether the candidate service at the current time t is available, whether the attribute value of the candidate service at the previous time is the same as that of the candidate service at the current time, and whether newly added candidate service resource exists;
When detecting that the service resource changes, the environment change response strategy comprises the following steps:
selecting a central point of the current moment PS and forming a representative individual set by extreme points on each objective function;
Calculating the distance from all individuals in the population to each representative individual, and associating the individual with the representative individual with the smallest distance;
predicting the population at the next moment according to the evolution step length of the individual represented at the current moment;
Randomly selecting individuals with set proportions in the predicted population to execute a diversity introduction strategy, and taking the obtained predicted population as an initial population under a new environment;
the diversity introduction strategy is: initializing individuals with set proportions in the prediction population;
PS is Pareto optimal solution set;
the solving process of the optimal service combination recommendation scheme comprises the following steps:
Performing service combination coding, namely coding a feasible service combination scheme into an individual in SLO, and constructing an initial population;
executing an environment change detection process, and when the environment change is identified, executing an environment change response strategy to form a prediction population under a new environment;
sequentially executing selection operation, cross operation and mutation operation in the micro-space aiming at the obtained predicted population, and selecting more optimal individuals to form the population obtained by operation in the micro-space;
Aiming at the population obtained by operation in the micro space, observation learning and imitation learning in the learning space are sequentially executed, and better individuals are selected to form the population obtained by optimization in the learning space;
aiming at the population obtained by optimization in the learning space, updating operation and influencing operation in the belief space are executed, and the population after the belief space optimization is obtained;
and iteratively executing the steps until the set iteration times are reached, outputting an optimal population, and obtaining an optimal solution set of the service combination.
5. An electronic device comprising a memory and a processor and computer instructions stored on the memory and running on the processor, which when executed by the processor, perform the steps of a dynamic multi-objective service composition optimization recommendation method according to any one of claims 1-3.
6. A computer readable storage medium storing computer instructions which, when executed by a processor, perform the steps of a dynamic multi-objective service composition optimization recommendation method of any one of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311388437.2A CN117349532B (en) | 2023-10-24 | 2023-10-24 | Dynamic multi-target service combination optimization recommendation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311388437.2A CN117349532B (en) | 2023-10-24 | 2023-10-24 | Dynamic multi-target service combination optimization recommendation method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117349532A CN117349532A (en) | 2024-01-05 |
| CN117349532B true CN117349532B (en) | 2024-05-24 |
Family
ID=89366430
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311388437.2A Active CN117349532B (en) | 2023-10-24 | 2023-10-24 | Dynamic multi-target service combination optimization recommendation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117349532B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106228232A (en) * | 2016-07-07 | 2016-12-14 | 淮北师范大学 | A kind of dynamic multi-objective based on fuzzy reasoning Population forecast strategy teaching optimization method |
| CN107306207A (en) * | 2017-05-31 | 2017-10-31 | 东南大学 | Calculated and multiple target intensified learning service combining method with reference to Skyline |
| CN114357724A (en) * | 2021-12-13 | 2022-04-15 | 中国人民解放军国防科技大学 | Dynamic multi-objective optimization-based opportunistic frequency planning method, device and equipment |
| CN114995964A (en) * | 2022-05-17 | 2022-09-02 | 烟台大学 | Combination service reconstruction method, device, equipment and computer readable medium |
| CN115470704A (en) * | 2022-09-16 | 2022-12-13 | 烟台大学 | Dynamic multi-objective optimization method, device, equipment and computer readable medium |
| CN116702633A (en) * | 2023-08-08 | 2023-09-05 | 北京理工大学 | Heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization |
| CN116738246A (en) * | 2023-06-12 | 2023-09-12 | 烟台大学 | Combined service dynamic reconstruction method and system for service demand change |
| CN116842843A (en) * | 2023-07-11 | 2023-10-03 | 郑州轻工业大学 | Dynamic multi-objective optimization method for irregular change of objective number |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120053970A1 (en) * | 2010-08-25 | 2012-03-01 | International Business Machines Corporation | Systems and methods for dynamic composition of business processes |
-
2023
- 2023-10-24 CN CN202311388437.2A patent/CN117349532B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106228232A (en) * | 2016-07-07 | 2016-12-14 | 淮北师范大学 | A kind of dynamic multi-objective based on fuzzy reasoning Population forecast strategy teaching optimization method |
| CN107306207A (en) * | 2017-05-31 | 2017-10-31 | 东南大学 | Calculated and multiple target intensified learning service combining method with reference to Skyline |
| CN114357724A (en) * | 2021-12-13 | 2022-04-15 | 中国人民解放军国防科技大学 | Dynamic multi-objective optimization-based opportunistic frequency planning method, device and equipment |
| CN114995964A (en) * | 2022-05-17 | 2022-09-02 | 烟台大学 | Combination service reconstruction method, device, equipment and computer readable medium |
| CN115470704A (en) * | 2022-09-16 | 2022-12-13 | 烟台大学 | Dynamic multi-objective optimization method, device, equipment and computer readable medium |
| CN116738246A (en) * | 2023-06-12 | 2023-09-12 | 烟台大学 | Combined service dynamic reconstruction method and system for service demand change |
| CN116842843A (en) * | 2023-07-11 | 2023-10-03 | 郑州轻工业大学 | Dynamic multi-objective optimization method for irregular change of objective number |
| CN116702633A (en) * | 2023-08-08 | 2023-09-05 | 北京理工大学 | Heterogeneous warhead task reliability planning method based on multi-objective dynamic optimization |
Non-Patent Citations (5)
| Title |
|---|
| 云计算环境中服务动态选择算法研究;张恒巍;韩继红;寇广;卫波;;计算机科学;20150515(第05期);全文 * |
| 基于多目标蚁群优化的知识即服务动态组合策略;贾瑞玉;伍章俊;张以文;;华南理工大学学报(自然科学版);20120615(第06期);全文 * |
| 基于混合预测策略与改进社会学习优化算法的动态多目标优化方法;张杰 等;《计算机应用研究》;20230405;第40卷(第4期);全文 * |
| 基于社会学习算法的函数优化与云任务调度方法研究;秦靖萱;《中国优秀硕士学位论文全文数据库信息科技辑》;20181115(第11期);全文 * |
| 面向函数优化的社会学习优化算法;刘志中 等;《小型微型计算机系统》;20170531;第38卷(第5期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117349532A (en) | 2024-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Coello | Constraint-handling techniques used with evolutionary algorithms | |
| Alabool et al. | Harris hawks optimization: a comprehensive review of recent variants and applications | |
| Chiroma et al. | Bio-inspired computation: Recent development on the modifications of the cuckoo search algorithm | |
| Zhou et al. | Multiobjective evolutionary algorithms: A survey of the state of the art | |
| Bao et al. | PSO-MISMO modeling strategy for multistep-ahead time series prediction | |
| Wu et al. | An Improved Teaching‐Learning‐Based Optimization Algorithm with Reinforcement Learning Strategy for Solving Optimization Problems | |
| Uyar et al. | A new population based adaptive domination change mechanism for diploid genetic algorithms in dynamic environments | |
| Nebro et al. | Analysis of leader selection strategies in a multi-objective particle swarm optimizer | |
| CN107197006B (en) | Multi-constraint service selection method and device based on global QoS decomposition | |
| Kaveh et al. | A hybrid multi-objective optimization and decision making procedure for optimal design of truss structures | |
| CN115470704B (en) | Dynamic multi-objective optimization method, device, equipment and computer readable medium | |
| Datta et al. | Graph partitioning by multi-objective real-valued metaheuristics: A comparative study | |
| Turky et al. | A dual-population multi operators harmony search algorithm for dynamic optimization problems | |
| Dantas et al. | On the use of fitness landscape features in meta-learning based algorithm selection for the quadratic assignment problem | |
| Cai et al. | Evolutionary approaches for multi-objective next release problem | |
| Coelho et al. | An immune-inspired multi-objective approach to the reconstruction of phylogenetic trees | |
| Seethalakshmi et al. | Real-coded multi-objective genetic algorithm with effective queuing model for efficient job scheduling in heterogeneous Hadoop environment | |
| Rezapoor Mirsaleh et al. | A learning automata-based memetic algorithm | |
| Zhang et al. | Reinforcement learning with actor-critic for knowledge graph reasoning | |
| CN117349532B (en) | Dynamic multi-target service combination optimization recommendation method and system | |
| Feng et al. | A dynamic opposite learning assisted grasshopper optimization algorithm for the flexible jobscheduling problem | |
| Ouyang et al. | Amended harmony search algorithm with perturbation strategy for large-scale system reliability problems | |
| Villacorta et al. | Sensitivity analysis in the scenario method: A multi-objective approach | |
| Annicchiarico et al. | Improved dynamical particle swarm optimization method for structural dynamics | |
| 崔逊学 et al. | A preference-based multi-objective concordance genetic algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |