A service composition optimization method in a collaborative service platform

Abstract

This invention proposes a service composition optimization method for a collaborative service platform, comprising: matching a set of candidate services that meet functional requirements based on user needs; standardizing the service quality attributes within the candidate service set and constructing a service composition optimization model based on the standardized service quality attributes; calculating the optimal solution with the highest fitness value in the service composition optimization model based on an improved flower pollination algorithm; and obtaining the optimal service composition of the service composition optimization mathematical model based on the optimal solution. The service composition optimization method for a collaborative service platform proposed in this invention improves the execution efficiency of service composition optimization.

Description

Technical Field

[0001] This invention belongs to the field of big data and artificial intelligence. Background Technology

[0002] Currently, cross-platform technology resource integration and service collaboration platforms need to aggregate large-scale technology resources and services. Each technology service is often optimized from multiple atomic services. The system first searches for candidate services based on service requirements to match atomic services, and then obtains the optimal service combination mainly based on conditions such as service quality, forming the final service solution. In this scenario, the problem to be solved is an NP-hard problem. For example, given that n = 5 atomic services are abstracted according to service requirements, and the number of candidate services corresponding to each atomic service is 1000, then the number of feasible solutions is 1000. 5 .

[0003] Therefore, improving the optimization algorithm is an effective way to improve the execution efficiency of service composition optimization.

[0004] This invention proposes a method to transform the service composition optimization problem into a polynomial in a mathematical model to find the optimal solution. Based on the classic FPA algorithm, a service composition optimization method for QoS-aware collaborative service platforms is proposed: a hybrid optimization multi-population pollination algorithm. This method first initializes the population with a set of optimal points to enhance population diversity; it then compares randomly generated numbers with P to adjust the pollination method; to improve the search capability of the global pollination process, it integrates a nonlinear adaptive weight ω and an exponentially decaying step size step(t) to dynamically adjust the search range; to improve the development capability of the local pollination process, it introduces elite solution information; simultaneously, it divides the entire population into multiple subpopulations using the mean position of individuals as the center point, and uses search strategies with different characteristics for collaborative search. Summary of the Invention

[0005] The present invention aims to at least partially solve one of the technical problems in the related art.

[0006] Therefore, the purpose of this invention is to propose a service composition optimization method in a collaborative service platform to improve the execution efficiency of service composition optimization.

[0007] To achieve the above objectives, a first aspect of the present invention proposes a service composition optimization method in a collaborative service platform, comprising:

[0008] Match a set of candidate services that meet the user's functional requirements;

[0009] The service quality attributes within the candidate service set are standardized, and a service composition optimization model is constructed based on the standardized service quality attributes.

[0010] The optimal solution with the highest fitness value in the service composition optimization model is calculated based on the improved flower pollination algorithm.

[0011] The optimal service combination is obtained from the optimal solution using the mathematical model for optimizing the service combination.

[0012] In addition, the service composition optimization method in a collaborative service platform according to the above embodiments of the present invention may also have the following additional technical features:

[0013] Furthermore, in one embodiment of the present invention, the standardization of service quality attributes within the candidate service set includes:

[0014] The attribute values ​​of different services are standardized to real numbers between 0 and 1.

[0015] For the i-th positive attribute value:

[0016]

[0017] For the i-th negative attribute value:

[0018]

[0019] in, and These are the maximum and minimum values ​​of the i-th attribute of the service.

[0020] Furthermore, in one embodiment of the present invention, the service composition optimization model includes:

[0021]

[0022] Among them, QoS cs For a certain QoS attribute of the combined service, C cs For the constraints of this QoS attribute, ω r The weights are for different QoS metrics, and satisfy the following conditions:

[0023] Furthermore, in one embodiment of the present invention, the improved flower pollination algorithm includes:

[0024] Adaptive adjustments are made to the pollination method;

[0025] Introduce nonlinear dynamic inertia weighting coefficients and dynamic step size;

[0026] Combined with information from elite solutions.

[0027] Furthermore, in one embodiment of the present invention, the adaptive adjustment of the pollination method includes:

[0028]

[0029] Where fitness(bestindex) is the best fitness value in the current population, and fitness(i) is the fitness value of the i-th individual.

[0030] Furthermore, in one embodiment of the present invention, the introduction of nonlinear dynamic inertia weighting coefficients and dynamic step size includes:

[0031] Introducing a nonlinear dynamic inertia weighting coefficient ω, the formula is as follows:

[0032]

[0033] Where, ω max and ω min Let ω represent the maximum and minimum values, respectively, and fitness(i) represent the current objective function value of the individual. avg and fitness min These represent the average target value and the minimum target value for all individuals, respectively.

[0034] Introducing the dynamic step size step(t), the formula is as follows:

[0035]

[0036] Where step(t) represents the current step size, t represents the number of iterations, and fitness i This represents the current individual fitness function value.

[0037] Furthermore, in one embodiment of the invention, the combination with the information of the elite solution includes:

[0038]

[0039] in, and This represents an individual randomly selected from the current subpopulation; This means randomly selecting one individual from the top 20% of the entire population; ε represents the current global optimal individual, and ε represents a random number distributed between [0, 1].

[0040] To achieve the above objectives, a second aspect of the present invention provides a service composition optimization device in a collaborative service platform, comprising the following modules:

[0041] The matching module is used to match a set of candidate services that meet the user's functional requirements.

[0042] The construction module is used to standardize the service quality attributes within the candidate service set and construct a service composition optimization model based on the standardized service quality attributes.

[0043] The calculation module is used to calculate the optimal solution with the highest fitness value of the service composition optimization model based on the improved flower pollination algorithm;

[0044] The acquisition module is used to acquire the optimal service combination of the service combination optimization mathematical model based on the optimal solution.

[0045] To achieve the above objectives, a third aspect of the present invention provides a computer device, characterized in that it includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements a service composition optimization method in a collaborative service platform as described above.

[0046] To achieve the above objectives, a fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements a service composition optimization method in a collaborative service platform as described above.

[0047] The service composition optimization method in the collaborative service platform proposed in this invention matches a set of candidate services that meet user needs and functional requirements; standardizes the service quality attributes within the candidate service set and constructs a service composition optimization model based on the standardized service quality attributes; calculates the optimal solution with the highest fitness value in the service composition optimization model based on an improved pollination algorithm; and obtains the optimal service composition of the service composition optimization mathematical model based on the optimal solution. By utilizing the optimized pollination algorithm, the efficiency of service composition optimization is improved. Compared to existing technologies, the advantages of this invention are that it represents QoS using an uncertain method and sets weights from both subjective and objective perspectives to express the uncertainty of QoS in real-world scenarios. Furthermore, it improves algorithm efficiency through solution space preprocessing and optimization algorithm adjustments and improvements. Attached Figure Description

[0048] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0049] Figure 1 This is a schematic diagram of a service composition optimization method in a collaborative service platform provided by an embodiment of the present invention.

[0050] Figure 2 This is a flowchart of a service composition optimization method in a collaborative service platform provided by an embodiment of the present invention.

[0051] Figure 3 This is a schematic diagram of an improved flower pollination algorithm in a collaborative service platform provided in an embodiment of the present invention.

[0052] Figure 4 This is a schematic diagram of a task workflow in a collaborative service platform provided in an embodiment of the present invention.

[0053] Figure 5 This is a schematic diagram of a service composition optimization device in a collaborative service platform provided in an embodiment of the present invention. Detailed Implementation

[0054] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0055] The service composition optimization method in the collaborative service platform of this invention is described below with reference to the accompanying drawings.

[0056] Web service discovery is the process of finding suitable web services based on certain criteria. How users can find the most suitable service among numerous options has become a hot research topic. Quality of Service (QoS) is a measure of a web service's ability to meet the needs of service requesters. The QoS model is an extensible vector that can describe service quality from many aspects, such as scalability, concurrency, response time, reliability, service price, throughput, availability, security, accuracy, and compensation rate, each evaluating service quality from different perspectives. In this invention, the following four QoS attributes are selected: Service Price (C), Response Time (RT), Availability (A), and Reliability (R).

[0057] Figure 1 This is a flowchart illustrating a service composition optimization method in a collaborative service platform provided by an embodiment of the present invention.

[0058] like Figure 1 As shown, the service composition optimization method in this collaborative service platform includes the following steps:

[0059] S101: Match a set of candidate services that meet the functional requirements based on user needs;

[0060] S102: Standardize the service quality attributes within the candidate service set, and construct a service composition optimization model based on the standardized service quality attributes;

[0061] S103: Calculate the optimal solution with the maximum fitness value in the service composition optimization model based on the improved flower pollination algorithm;

[0062] S104: Obtain the optimal service composition from the mathematical model of service composition optimization based on the optimal solution.

[0063] Figure 2 This is a flowchart of the service composition optimization method in the collaborative service platform of this invention. Service attributes are generally divided into positive attributes and negative attributes. For positive attributes, the larger the value, the higher the quality of service provided, such as service capacity and throughput. For negative attributes, the smaller the value, the higher the quality of service provided, such as service cost and time. To eliminate the characteristic differences between different attributes, we standardize the attribute values ​​of different services to real numbers between 0 and 1.

[0064] Furthermore, in one embodiment of the present invention, the service quality attributes within the candidate service set are standardized, including:

[0065] The attribute values ​​of different services are standardized to real numbers between 0 and 1.

[0066] For the i-th positive attribute value:

[0067]

[0068] For the i-th negative attribute value:

[0069]

[0070] in, and These are the maximum and minimum values ​​of the i-th attribute of the service.

[0071] Furthermore, in one embodiment of the present invention, the service composition optimization model includes:

[0072]

[0073] Among them, QoS cs For a certain QoS attribute of the combined service, C cs For the constraints of this QoS attribute, ω r The weights are for different QoS metrics, and satisfy the following conditions:

[0074] The two main determinants of the efficiency of intelligent algorithms are: development capability and search capability.

[0075] To resolve the conflict between algorithm development capability and exploration capability, the classic FPA algorithm introduces a transformation probability p. FPA performs global pollination with probability p and local pollination with probability 1-p. Therefore, the transformation probability p is of great significance to the FPA algorithm.

[0076] like Figure 3 The diagram shows the improved flower pollination algorithm flow in the collaborative service platform of this invention.

[0077] Furthermore, in one embodiment of the present invention, the improved flower pollination algorithm includes:

[0078] Adaptive adjustments are made to the pollination method;

[0079] Introduce nonlinear dynamic inertia weighting coefficients and dynamic step size;

[0080] Combined with information from elite solutions.

[0081] In the basic flower pollination algorithm, p is a constant, meaning that the probability of performing global pollination and local pollination operations remains unchanged during algorithm execution. However, if the value of p is too large, the number of global pollination operations will be too high, which may lead to a global optimum but not convergence; if the value of p is too small, the number of local pollination operations will be too high, which may lead to getting trapped in a local optimum. Therefore, an adaptive adjustment is made to the pollination method.

[0082] Furthermore, in one embodiment of the present invention, adaptive adjustments are made to the pollination method, including:

[0083]

[0084] Where fitness(bestindex) is the best fitness value in the current population, and fitness(i) is the fitness value of the i-th individual.

[0085] If P(T, i) < P0, then global pollination is performed; otherwise, local pollination is performed. The improved algorithm can adaptively adjust the execution probability of global search and local search, effectively solving the balance problem between the algorithm's development capability and exploration capability. It can not only avoid the FPA algorithm from getting trapped in local optima, but also improve the convergence speed of the algorithm.

[0086] To enhance the development capabilities of FPA, inspired by the Adaptive Particle Swarm Optimization (APSO) algorithm, a nonlinear dynamic inertia weight coefficient is introduced. To further enhance the search capability of FPA, a dynamic step size is introduced.

[0087] Furthermore, in one embodiment of the present invention, nonlinear dynamic inertia weighting coefficients and dynamic step sizes are introduced, including:

[0088] Introducing a nonlinear dynamic inertia weighting coefficient ω, the formula is as follows:

[0089]

[0090] Where, ω max and ω min Let ω represent the maximum and minimum values, respectively, and fitness(i) represent the current objective function value of the individual. avg and fitness min These represent the average target value and the minimum target value for all individuals, respectively.

[0091] Specifically, by adjusting the value of ω, the system aims to escape local minima (local optima). When the objective function values ​​of all individuals tend to be consistent or approach a local optimum, the inertia weight is increased. Conversely, when the objective function values ​​of individuals are relatively dispersed, the inertia weight is decreased. Furthermore, for individuals whose objective function values ​​are better than the average objective function value, their corresponding inertia weight factor is smaller, thus preserving that individual. On the other hand, for individuals whose objective function values ​​are worse than the average objective function value, their corresponding inertia weight factor is larger, causing that individual to move closer to a better search region.

[0092] Introducing the dynamic step size step(t), the formula is as follows:

[0093]

[0094] Where step(t) represents the current step size, t represents the number of iterations, and fitness i This represents the current individual fitness function value.

[0095] Specifically, the algorithm uses a decreasing step size to escape local minima (local optima). If self-pollination occurs in the early stages of iteration, a larger step size factor gives the algorithm a better ability to escape local optima in the early stages. As the number of iterations increases, the algorithm mainly performs self-pollination in the later stages, and the accuracy of the step size factor also increases by orders of magnitude, effectively improving the algorithm's solution accuracy.

[0096] To improve the algorithm's local search capability, i.e., its exploitation ability, it is combined with information from elite solutions.

[0097] Furthermore, in one embodiment of the invention, the information combined with the elite solution includes:

[0098]

[0099] in, and This represents an individual randomly selected from the current subpopulation; This means randomly selecting one individual from the top 20% of the entire population; ε represents the current global optimal individual, and ε represents a random number distributed between [0, 1].

[0100] The above is a complete process for optimizing service composition in a collaborative service platform. Figure 4 This is a schematic diagram of the task workflow in the collaborative service platform of this invention. Typically, a workflow contains many tasks representing user needs, each task being an abstraction of a user need, denoted by AS. n represents the number of tasks, WS represents the services available to complete a given task, and m is its number. In reality, the value of m varies, but in this invention, for simplicity, the variables for each dataset are fixed.

[0101] The service composition optimization method in the collaborative service platform proposed in this invention matches a set of candidate services that meet user needs and functional requirements; standardizes the service quality attributes within the candidate service set and constructs a service composition optimization model based on the standardized service quality attributes; calculates the optimal solution with the highest fitness value in the service composition optimization model based on an improved pollination algorithm; and obtains the optimal service composition of the service composition optimization mathematical model based on the optimal solution. By utilizing the optimized pollination algorithm, the efficiency of service composition optimization is improved. Compared to existing technologies, the advantages of this invention are that it represents QoS using an uncertain method and sets weights from both subjective and objective perspectives to express the uncertainty of QoS in real-world scenarios. Furthermore, it improves algorithm efficiency through solution space preprocessing and optimization algorithm adjustments and improvements.

[0102] To implement the above embodiments, the present invention also proposes a service composition optimization device in a collaborative service platform.

[0103] Figure 5 This is a schematic diagram of the structure of a service composition optimization device in a collaborative service platform provided in an embodiment of the present invention.

[0104] like Figure 5 As shown, the service composition optimization device in this collaborative service platform includes: a matching module 100, a construction module 200, a calculation module 300, and an acquisition module 400, wherein...

[0105] The matching module is used to match a set of candidate services that meet the user's functional requirements.

[0106] The construction module is used to standardize the service quality attributes within the candidate service set and construct a service composition optimization model based on the standardized service quality attributes.

[0107] The calculation module is used to calculate the optimal solution with the highest fitness value of the service composition optimization model based on the improved flower pollination algorithm;

[0108] The acquisition module is used to acquire the optimal service combination of the service combination optimization mathematical model based on the optimal solution.

[0109] To achieve the above objectives, a third aspect of the present invention provides a computer device, characterized in that it includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the service composition optimization method in the collaborative service platform as described above.

[0110] To achieve the above objectives, a fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the service composition optimization method in the collaborative service platform as described above.

[0111] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0112] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0113] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A service composition optimization method in a collaborative service platform, characterized in that, The method comprises the following steps: matching a candidate service set meeting the functional requirement according to the user requirement; standardizing the service quality attributes in the candidate service set and constructing a service combination optimization model according to the standardized service quality attributes; calculating an optimal solution with the maximum fitness value of the service combination optimization model based on an improved flower pollination algorithm; obtaining an optimal service combination of the service combination optimization mathematical model according to the optimal solution; wherein the improved flower pollination algorithm comprises: self-adaptive adjustment of the pollination mode; introduction of a nonlinear dynamic inertia weight coefficient and a dynamic step length; combination of information of an elite solution; the self-adaptive adjustment of the pollination mode comprises: in, This represents the best fitness value for the current population. For the first If the fitness value of an individual is P(T,i) < P0, then global pollination is performed; otherwise, local pollination is performed. the combination of information of the elite solution comprises: ， where, and represents an individual randomly selected from the current sub-population; represents an individual randomly selected from the top 20% of the entire population; represents the current global best individual, represents a random number distributed in [0, 1], represents a non-linear dynamic inertia weight coefficient.

2. The method of claim 1, wherein, the standardization of the service quality attributes in the candidate service set comprises: standardizing attribute values of different services into real numbers between 0 and 1, For the first positive attribute value: For the first negative attribute value: wherein, and are the maximum and minimum values of the service the th attribute.

3. A service combination optimization apparatus in a service platform, characterized by, The method comprises the following steps: a matching module, configured to match a candidate service set meeting the functional requirement according to the user requirement; a constructing module, configured to standardize the service quality attributes in the candidate service set and construct a service combination optimization model according to the standardized service quality attributes; a calculating module, configured to calculate an optimal solution with the maximum fitness value of the service combination optimization model based on an improved flower pollination algorithm; an obtaining module, configured to obtain an optimal service combination of the service combination optimization mathematical model according to the optimal solution; wherein the improved flower pollination algorithm comprises: self-adaptive adjustment of the pollination mode; introduction of a nonlinear dynamic inertia weight coefficient and a dynamic step length; combination of information of an elite solution; the self-adaptive adjustment of the pollination mode comprises: in, This represents the best fitness value for the current population. For the first If the fitness value of an individual is P(T,i) < P0, then global pollination is performed; otherwise, local pollination is performed. the combination of information of the elite solution comprises: ， where, and represents an individual randomly selected from the current sub-population; represents an individual randomly selected from the top 20% of the entire population; represents the current global best individual, represents a random number distributed in [0, 1], represents a nonlinear dynamic inertia weight coefficient.

4. A computer device, comprising: A computer program product comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the service combination optimization method in the collaborative service platform according to any one of claims 1-2 when executing the computer program.

5. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program product is executed by the processor to implement the service combination optimization method in the collaborative service platform according to any one of claims 1-2.

Images