Mobile crowdsourcing strategy optimization method and system based on multi-objective optimization

Abstract

The invention relates to the crossing field of the crowd sensing technology and the artificial intelligence technology, in particular to a mobile crowdsourcing strategy optimization method and system based on multi-objective optimization, and the method comprises the steps: carrying out the task information publishing through a crowdsourcing platform, and enabling the task information to comprise the publishing task type, the highest budget and the equipment requirement; the method comprises the steps of performing information pushing according to a release position of task information, calculating a worker information attribute based on the released task information and uploading the worker information attribute to a crowdsourcing platform, performing task-worker matching according to the worker information attribute and the task information, and performing real-time dynamic pricing according to actual conditions of a current task and a worker by using an initial pricing model. According to the method, in combination with the task completion number and the current reputation value of workers on the day and the current reputation and the current score of an encryption mechanism, all the workers uploading information are preliminarily screened to obtain a worker candidate list, an initial solution set is generated, the search range is narrowed through the process, and the unnecessary calculation amount is reduced.

Description

Technical Field

[0001] The present invention relates to the intersection of crowd intelligence perception technology and artificial intelligence technology, and in particular to a mobile crowdsourcing strategy optimization method and system based on multi-objective optimization. Background Art

[0002] Mobile crowd sensing is an example of collective intelligence achieved through multiple sensors and devices in the IoT ecosystem. Supported by IoT devices, edge computing, and drone sensing, crowd sensing plays a vital role in urban sensing, environmental monitoring, and traffic management. In the field of mobile crowd sensing research, recruiting the right workers to complete tasks and allocating benefits to all parties (i.e., pricing issues) are two key challenges. Existing recruitment strategies are usually driven by efficiency, quality, and coverage, but ignore the matching between workers and tasks, which directly affects the quality of task completion. In addition, existing pricing methods mostly use tiered pricing, difficulty pricing, and fixed pricing, which do not fully consider the differences between the historical status and current status of all parties, and cannot fully motivate the participation enthusiasm of all parties, thus affecting the overall profit and participation.

[0003] Traditional mobile crowd intelligence perception systems have defects such as weak privacy protection, low worker selection efficiency, lack of dynamic incentive mechanism and unreasonable profit distribution. The platform bears the dual responsibilities of data encryption and worker recruitment, resulting in insufficient privacy protection. The worker selection mechanism cannot effectively cope with the complexity of multi-tasks and is prone to fall into local optimality. The fixed incentive mechanism lacks flexibility and cannot motivate workers to participate in tasks with high quality. In addition, the profit distribution method is unreasonable and the interests of all parties have not been effectively balanced. At this stage, a mobile crowdsourcing strategy optimization method and system based on multi-objective optimization is needed. Summary of the invention

[0004] In order to solve the problems of weak privacy protection, low worker selection efficiency, lack of dynamic incentive mechanism and unreasonable benefit distribution in traditional mobile crowdsourcing perception systems, the present invention provides a mobile crowdsourcing strategy optimization method and system based on multi-objective optimization, which can achieve accurate matching between workers and tasks and efficient worker recruitment, maximize the completion effect of tasks, and ensure reasonable pricing to maximize the interests of all parties.

[0005] In a first aspect, the present invention provides a mobile crowdsourcing strategy optimization method based on multi-objective optimization, which adopts the following technical solution: A mobile crowdsourcing strategy optimization method based on multi-objective optimization, comprising: Use crowdsourcing platforms to publish task information, including task type, maximum budget, and equipment requirements; Pushing information based on the location where the task information is published, including extracting the geographical location where the task information is published, and pushing the task based on a preset broadcast range; Calculate the worker information attributes based on the published task information and upload them to the crowdsourcing platform, where the calculated worker information includes the distance attribute, reputation attribute, minimum income and capability value attribute of the calculation task; Perform task-worker matching based on worker information attributes and task information, including using the EH-STLS algorithm to perform task-worker matching and generate an optimal task allocation information table; The optimal task allocation information table is input into the KDQN dynamic pricing model, and dynamic pricing is performed through the Stackelberg game to obtain the initial pricing model; The initial pricing model is used to conduct real-time dynamic pricing based on the current tasks and the actual situation of the workers, including combining the number of tasks completed by the workers on that day, the current reputation value, the current reputation of the encryption agency, and the current rating.

[0006] Furthermore, the worker information attributes are calculated based on the published task information and uploaded to the crowdsourcing platform, including calculating the distance attribute according to the geographic information in the task information and the current geographic information of the worker, and normalizing the calculated distance attribute. The distance attribute calculation formula is: , in, is the radius of the Earth, is the dimensional difference between the worker's location and the task's posting location, is the dimension of the worker’s current geographical information, is the dimension of the task publishing location, is the longitude difference between the worker's location and the task's posting location.

[0007] Furthermore, the use of the EH-STLS algorithm for task-worker matching includes preliminary screening of worker information attributes of all uploaded information based on task information and worker information attributes to obtain a list of worker candidates, and generating an initial solution set based on the list of worker candidates, and using the EH-STLS algorithm to perform a comprehensive evaluation in the initial solution set, so that each task is matched with only one worker at this stage, thereby generating a preliminary global optimal solution set.

[0008] Furthermore, the use of the EH-STLS algorithm for task-worker matching also includes performing dynamic crossover and mutation iterations based on the initial solution set and satisfying requirements and constraints, and further searching for high-quality Pareto solutions through variable domain local search after a certain number of iterations to generate an optimal task allocation information table.

[0009] Furthermore, the inputting of the optimal task allocation information table into the KDQN dynamic pricing model includes parsing and preprocessing the data in the task allocation information table, and constructing a task-worker allocation matrix, then extracting the task features and detailed attributes of the workers in the task allocation information table, and normalizing the extracted numerical data.

[0010] Furthermore, the dynamic pricing operation is performed through the Stackelberg game to obtain an initial pricing model, including using the reverse induction method to solve the optimal strategy of followers, solving the optimal work effort and pricing strategy of workers and encryption agencies respectively, and then solving the optimal pricing strategy of the platform, and then determining the Stackelberg equilibrium solution, constructing the initial pricing model based on the equilibrium solution, and obtaining the model function by fitting the relationship between the pricing strategy and the task requirements and budget.

[0011] Furthermore, the initial pricing model is used to perform real-time dynamic pricing based on the actual situation of the current task and the worker, including obtaining the number of tasks completed on the day and the current reputation value in real time for the worker, and integrating them with the worker ability information in the initial pricing model into a real-time state vector for the worker. For the task, the task type, maximum budget and equipment requirements in the initial pricing model are combined to form a real-time state vector for the task, and finally summarized into a complete real-time state vector for pricing calculation.

[0012] In the second aspect, a mobile crowdsourcing strategy optimization system based on multi-objective optimization includes: The data acquisition module is configured to: utilize the crowdsourcing platform to publish task information, including publishing task type, maximum budget and equipment requirements; The push module is configured to: push information according to the publishing location of the task information, including extracting the geographical location where the task information is published, and push the task according to a preset broadcast range; The attribute module is configured to: calculate the worker information attributes based on the published task information and upload the attributes to the crowdsourcing platform, wherein the calculated worker information includes calculating the distance attribute, reputation attribute, minimum income and capability value attribute of the task; The feature extraction module is configured to: perform task-worker matching according to worker information attributes and task information, including performing task-worker matching using an EH-STLS algorithm, and generating an optimal task allocation information table; The model module is configured to: input the optimal task allocation information table into the KDQN dynamic pricing model, perform dynamic pricing operation through the Stackelberg game, and obtain an initial pricing model; The transformation module is configured to: use the initial pricing model to perform real-time dynamic pricing based on the current task and the actual situation of the worker, including combining the number of tasks completed by the worker on the day, the current reputation value, and the encryption agency's current reputation and current score.

[0013] In a third aspect, the present invention provides a computer-readable storage medium storing a plurality of instructions, wherein the instructions are suitable for being loaded and executed by a processor of a terminal device for the mobile crowdsourcing strategy optimization method based on multi-objective optimization.

[0014] In a fourth aspect, the present invention provides a terminal device comprising a processor and a computer-readable storage medium, wherein the processor is used to implement various instructions; the computer-readable storage medium is used to store multiple instructions, wherein the instructions are suitable for being loaded by the processor and executing the mobile crowdsourcing strategy optimization method based on multi-objective optimization.

[0015] In summary, the present invention has the following beneficial technical effects: 1. The present invention uses an enhanced heuristic search algorithm based on taboo and local search (EH-STLS) to match tasks and workers, comprehensively considers the task type, budget and various constraints, and can accurately screen out the most suitable workers for each task. According to factors such as task location, equipment requirements, worker reputation and ability, the task is assigned to workers in the area who have high-precision monitoring equipment and a good work record, which greatly improves the accuracy of task allocation.

[0016] 2. In the EH-STLS algorithm of the present invention, all workers who uploaded information were preliminarily screened to obtain a list of worker candidates and generate an initial solution set. This process narrows the search scope and reduces unnecessary calculations. At the same time, the algorithm is used to perform a comprehensive evaluation in the initial solution set. Each task is matched with only one worker to generate a preliminary global optimal solution set, which further focuses the search direction.

[0017] 3. In the KDQN model, the present invention parses and preprocesses the data in the task allocation information table, extracts task features and detailed attributes of workers, and normalizes the numerical data, thereby providing an accurate and standardized data basis for the pricing model.

[0018] 4. The present invention uses the reverse induction method to solve the optimal strategy of followers in the Stackelberg game, determine the Stackelberg equilibrium solution and construct an initial pricing model. The model function is obtained by fitting the relationship between the pricing strategy and the task requirements and budget, so that the pricing model has a scientific theoretical basis and strong adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is an overall system framework diagram of a mobile crowdsourcing strategy optimization method based on multi-objective optimization according to Embodiment 1 of the present invention; Figure 2A comparison diagram of the KAN layer and the MLP layer of the reinforcement learning dynamic pricing model based on Stackelberg game in Example 1 of the present invention; Figure 3 This is a graph showing the experimental results of the task quality of workers according to Example 1 of the present invention as the number of tasks changes; Figure 4 This is a graph showing the results of a task quality experiment in which the number of workers changes with the task in Example 1 of the present invention; Figure 5 This is a pricing and benefit diagram for different numbers of tasks in Example 1 of the present invention. DETAILED DESCRIPTION

[0020] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0021] Example 1 Reference Figure 1 , a mobile crowdsourcing strategy optimization method based on multi-objective optimization in this embodiment includes: Use crowdsourcing platforms to publish task information, including task type, maximum budget, and equipment requirements; Pushing information based on the location where the task information is published, including extracting the geographical location where the task information is published, and pushing the task based on a preset broadcast range; Calculate the worker information attributes based on the published task information and upload them to the crowdsourcing platform, where the calculated worker information includes the distance attribute, reputation attribute, minimum income and capability value attribute of the calculation task; Perform task-worker matching based on worker information attributes and task information, including using the EH-STLS algorithm to perform task-worker matching and generate an optimal task allocation information table; The optimal task allocation information table is input into the KDQN dynamic pricing model, and dynamic pricing is performed through the Stackelberg game to obtain the initial pricing model; The initial pricing model is used to conduct real-time dynamic pricing based on the current tasks and the actual situation of the workers, including combining the number of tasks completed by the workers on that day, the current reputation value, the current reputation of the encryption agency, and the current rating.

[0022] Specifically, In the mobile crowd intelligence perception system, there are four main roles: task publisher, worker, platform and encryption agency; for example, consider a task that requires data collection issued by a task publisher, and then the platform broadcasts the task near the task location. After receiving the information uploaded by the worker, the platform selects the best worker through multi-objective optimization, and then the platform puts the four parties together for dynamic pricing, which can greatly guarantee the quality of task completion and the distribution of benefits for all parties involved.

[0023] like Figure 1As shown, the present invention designs a mobile crowdsourcing efficiency strategy optimization framework based on multi-objective optimization and reinforcement learning game method. In the framework of worker recruitment and dynamic pricing, worker recruitment is obtained by multi-objective optimization method, dynamic pricing is obtained by reinforcement learning game method, and task benefits and equipment similarity in worker recruitment are calculated according to the task requirements issued by the task publisher and the worker attributes uploaded by the worker. When calculating task benefits and equipment similarity, multiple ability indicators of workers and weight requirements of task publishers are considered. Subsequently, the platform considers the task benefits and equipment similarity of each worker for different tasks, and calculates an initial list of workers for each task. On this basis, this initial list of workers is used as the input of multi-objective optimization and put into the multi-objective optimization algorithm. At this time, the worker recruitment problem is modeled as a multi-objective optimization problem, that is, while ensuring the quality of task completion and task requirements, maximizing task benefits and maximizing satisfaction of task requirements. Dynamic pricing is to train an initial model based on the historical pricing results of workers, platforms, etc. On this basis, dynamic pricing is performed through a pricing model considering the number of tasks completed by workers on the day, the current reputation value, the current reputation of the encryption agency, the current score, etc. At this point, the dynamic pricing problem is modeled as a Stackelberg game problem, that is, the task publisher is the leader, and the platform, encryption agency and workers are followers. On the basis of ensuring the minimum benefits of all parties, the maximum benefit distribution is given to all parties. The specific steps are as follows: Embodiment 1 of the present invention provides an enhanced heuristic search method based on taboo and local search, including a task information broadcasting stage, a worker information uploading stage and a task-worker matching stage.

[0024] Mission information broadcast phase: Step 1: The task publisher sets the task type, maximum budget, and equipment requirements through the crowdsourcing platform and then publishes the task information.

[0025] Step 2: After receiving the task information, the crowdsourcing platform broadcasts the task information near the task point based on the geographic location in the task information.

[0026] Step 3: After broadcasting the task information, wait for the workers to upload their own information in the information uploading phase, and then proceed to the next phase.

[0027] Worker information upload stage: Step 4: Workers will judge the broadcasted tasks, and only after they determine the tasks they are interested in will they start calculating their own relevant information.

[0028] In step 4, the specific steps for workers to judge the broadcasted tasks are: Step 4.1: The worker compares the task type with his / her preference list as follows: , in, is the task type, Preference list for workers.

[0029] Step 4.2: Determine the result of the previous step. If the result is 1, perform the next calculation operation; otherwise, ignore the task information.

[0030] Step 5: The worker calculates the distance attribute for the task based on the geographic information in the task information and the geographical information of the worker's current location. The worker calculates the distance attribute for the task based on the geographic information in the task information and the geographical information of the worker's current location. The distance attribute calculation formula is: , in, is the radius of the Earth, is the dimensional difference between the worker's location and the task's posting location, is the dimension of the worker’s current geographical information, is the dimension of the task publishing location, is the longitude difference between the worker's location and the task's posting location.

[0031] Step 6: After calculating the distance attribute for the task, normalize it to facilitate subsequent task allocation.

[0032] Step 7: The worker will then calculate his or her own reputation attribute based on the completion of the last N tasks. The worker will calculate his or her own reputation attribute based on the completion of the last N tasks.

[0033] In this embodiment, the reputation attribute The calculation of is as follows: , in, is a reputation base value. is the average reputation of all workers, It is the sum of the reputation scores of the worker in the last N tasks.

[0034] Step 8: Workers also need to calculate their current minimum income as a reference to prevent the situation where income is lower than cost and protect the rights and interests of workers. Workers calculate their current minimum income as shown in the following formula: , in is the regional impact factor, which is determined by the local level and has a value between (0,1). is the profit obtained each time, and S is the number of historical records.

[0035] Step 9: Finally, the worker calculates the ability value attribute for the task of interest based on the task requirements and his own equipment. The worker calculates the ability value attribute for the task of interest based on the task requirements and his own equipment. The calculation is as follows: , in, is the reputation information of the worker, is the maximum working distance of the workers, , and Represent the weight coefficients of different items respectively. is the Jaccard similarity, which is used to evaluate the similarity between the worker's equipment situation and the equipment requirements for task completion, where It is expressed as: , in, Represents the set of equipment requirements needed to complete the task. This set includes the equipment required to complete a specific task. The characteristics and performance requirements of various equipment required, Indicates workers The equipment status collection covers the various characteristics of the equipment actually equipped by the worker.

[0036] Step 10: After the workers have calculated this information locally, they upload it to the crowdsourcing platform for the next task assignment operation.

[0037] Task-worker matching phase: Step 11: After receiving the attribute information from the workers, the crowdsourcing platform generates an initial set of workers for each task.

[0038] Step 12: The platform puts this set of workers into the EH-STLS algorithm, which considers the type of task, budget, and other relevant constraints to select the most suitable workers for each task.

[0039] The worker recruitment problem is modeled as a multi-objective optimization problem. The worker information and task information are processed using an enhanced heuristic search algorithm based on taboo and local search to obtain the task Best executive worker The specific steps are: Step 12.1: Categorize all the information and organize it according to different tasks. Each task corresponds to a separate list of worker candidates, and the system generates an initial solution set based on these candidate lists.

[0040] Step 12.2: In the initial solution set, each task has M candidate workers. Based on this, the algorithm will generate a preliminary global optimal solution set for each task, in which each task is matched with only one worker. Then, this preliminary solution set is added to the taboo table. The taboo table is used to record the solutions that have been searched to prevent the algorithm from repeatedly selecting the same solution in a short period of time.

[0041] Step 12.3: In the evolution process, crossover and mutation operations are combined to increase the solution exploration capability through a variety of operation mechanisms. This embodiment designs a probability strategy based on dynamic adjustment of the number of iterations, so that the crossover and mutation probabilities can change dynamically as the iterations proceed, as shown in the following formula: , , Among them, g is the current iteration number, N is the total number of iterations, is the crossover probability, is the mutation probability.

[0042] Step 12.4: After a certain number of iterations, a local search strategy with a variable neighborhood is introduced to deeply optimize the solution using different neighborhood structures to improve the convergence quality of the optimal solution and ultimately generate the optimal task allocation list. The specific formula is as follows: , ,

[0043] in, is the maximum neighborhood value, is the maximum mutation probability, is the minimum mutation probability, Pm(g) is the neighborhood coefficient of generation g, N(g) is the neighborhood size of generation g, T is the taboo table, It is the desire criterion. When the objective function value of a taboo solution is better than the currently known optimal solution, it is allowed to be selected as the new solution even if it is in the taboo table.

[0044] Step 13: Based on the screening results of the EH-STLS algorithm, the platform generates an optimal task allocation information table, which lists the correspondence between each task and its best worker.

[0045] Step 14: The platform inputs the optimal task allocation information table into the pricing model and proceeds to the next step of dynamic pricing operation.

[0046] The following is a detailed introduction to the worker recruitment problem through an example. Assuming that the task publisher submits 10 tasks, the platform will broadcast them according to the task locations of these 10 tasks. Taking into account the distance efficiency of workers, the platform sets the broadcast range to 50 kilometers near the task point. After the workers at each location receive the task information, they calculate the relevant information for the tasks that meet the requirements and upload it. After the platform receives the information uploaded by the workers, it generates a list of 10 candidate workers and puts these 10 candidate worker lists into the multi-objective optimization algorithm. After multiple rounds of iterations, while ensuring the quality of task completion and meeting the task requirements, the constraints are met at the same time, and finally an optimal worker matching list is obtained, which contains 10 workers and 10 tasks, and they correspond one to one.

[0047] In order to verify the effectiveness of the method of this embodiment, an enhanced heuristic search algorithm based on taboo and local search is experimented with a real data set, wherein the real data set is derived from existing public data and the consent of the information owner is obtained.

[0048] (1) Task quality, Among them, EH-STLS represents the multi-objective optimization algorithm proposed in this invention, and the other algorithms are comparative algorithms, namely SPEA2: It evaluates individual fitness by combining dominance relations and density estimation, and uses external archives to save the optimal solution, thereby ensuring the quality and diversity of the solution. NSGA-II: It retains high-quality individuals by using elite strategies, generates new solutions using crossover and mutation operations, and finally finds a set of balanced Pareto frontier solutions. MOEA-D: It solves the multi-objective optimization problem by decomposing it into several single-objective sub-problems. Each sub-problem is represented by a different weighted combination, and uses neighborhood information to collaborate with other sub-problems to improve search efficiency and solution diversity. MOGNDO: It saves non-dominated solutions and guides the optimization process by introducing external archives and leader selection mechanisms, ensuring the diversity and convergence of solutions.

[0049] like Figure 3 As shown in the figure, when the number of workers is fixed, as the number of tasks increases, the task quality of each algorithm shows a downward trend. This is because when the number of tasks increases, the system prioritizes ensuring that each task is assigned to a worker, thus partially compromising the task quality. The task quality performance of the EH-STLS algorithm when the number of tasks is 50, 100, and 150 is better than that of other comparison algorithms, especially compared with the MOEA-D algorithm, the task quality is improved by 6.6%, 11.1%, and 11.9%, respectively. This shows that the EH-STLS algorithm can more effectively maintain the execution quality of tasks when the task scale increases.

[0050] like Figure 4As shown in the figure, when the number of tasks is fixed, as the number of workers increases, the task quality of each algorithm shows an upward trend. This is because the increase in the number of workers gives the system more choices when allocating tasks, thereby improving the quality of tasks. The task quality performance of the EH-STLS algorithm when the number of tasks is 50, 100, and 150 is better than that of the other comparison algorithms, which is 21.7%, 13.3%, and 12.0% higher than that of the MOEA-D algorithm, respectively. This shows that EH-STLS can more effectively improve the efficiency of task allocation when the number of workers is small. Although the improvement decreases with the increase in the number of workers, EH-STLS still maintains a significant advantage.

[0051] When the number of workers is fixed, the task benefits of each algorithm show a downward trend as the number of tasks increases. This is because when the number of tasks increases, the system prioritizes ensuring that each task has a worker assigned, thus compromising on task benefits. However, the task benefits of the EH-STLS algorithm are higher than those of the comparison algorithm when the number of tasks is 50, 100, and 150, which are 2.8%, 5.4%, and 9.7% higher than those of the MOEA-D algorithm, respectively. This shows that as the size of the task increases, the advantage of EH-STLS in task benefits becomes more significant.

[0052] When the number of tasks is fixed, as the number of workers increases, the task benefits of each algorithm show an upward trend. This is because the increase in the number of workers provides the system with more options, thereby improving the task benefits. The task benefits of the EH-STLS algorithm when the number of workers is 200, 300, and 400 are higher than those of the other comparison algorithms, and are 4.3%, 5.6%, and 3.6% higher than those of the MOEA-D algorithm, respectively. This shows that when the number of workers is sufficient, EH-STLS has a stronger task allocation capability and can significantly improve the effectiveness of tasks.

[0053] By comparing the task quality and task efficiency of different algorithms under different numbers of workers and different numbers of tasks, the performance of the EH-STLS algorithm is better than other algorithms. This is because the EH-STLS algorithm uses a higher mutation probability in the early stage of the search to increase the diversity of solutions and enhance the ability of the algorithm to jump out of the local optimum; and as the number of iterations increases, the mutation probability is gradually reduced to enhance the deep search capability of the current solution and ensure convergence to a better global optimal solution. Through this dynamic adjustment mechanism, a balance between exploration and development can be achieved between different stages, thereby effectively improving the overall quality and efficiency of task allocation. Secondly, in order to further improve the quality of the solution, a local search strategy with a variable neighborhood is introduced after a certain number of iterations, and different neighborhood structures are used to deeply optimize the solution to improve the convergence quality of the optimal solution.

[0054] Extensive experiments are conducted on three real datasets to verify the performance of the proposed enhanced heuristic search algorithm based on taboo and local search. The results show that the enhanced heuristic search algorithm based on taboo and local search outperforms other baselines.

[0055] In the present invention, the Kolmogorov-Arnold network (KAN) is used to replace the MLP layer in DQN. Unlike MLP, KAN does not use fixed activation functions (such as ReLU, Sigmoid, etc.). Instead, it uses learnable activation functions on "weights", which are parameterized by splines, so that more complex function mappings can be learned during training. MLP learns features through alternating combinations of linear weight matrices (fully connected layers) and nonlinear activation functions, while KAN does not have such linear weight matrices. Instead, each weight parameter is replaced by a single variable learnable function (usually a spline function), which makes KAN outperform MLP in low-dimensional data fitting and partial differential equation (PDE) solving. In KAN, nodes only perform simple summation operations on incoming signals without applying any nonlinearity, and all nonlinearities are performed on the edges.

[0056] like Figure 2 As shown in Figure 1, KAN is introduced to replace MLP. KAN uses a learnable activation function, which makes it more flexible in approximating complex functional relationships, so that it can better fit the Q value. At this time, the Q function is changed to: , in, is a learnable activation function, are the parameters of KAN.

[0057] In addition, the structure of KAN enables it to adjust the form of the activation function according to the complexity of the input data during training, thereby improving the convergence speed of the model. The convergence speed is improved to: , in, It represents the convergence rate index related to the number of model parameters N. Expressed as the convergence speed of the KAN network, In KAN, the activation function itself is also a part of the network, which is continuously learned and adjusted through the optimization process. This allows the activation function of each layer to be adaptively adjusted according to the characteristics of the data, thereby achieving more flexible nonlinear mapping. The output function of KAN is expressed as: , in, It is based on the input state s and action a and model parameters To calculate an estimate, is a univariate learnable function, is a global nonlinear combination function.

[0058] This means that the model can fine-tune each input feature Impact on output,Compared with MLP, KAN provides richer derivative forms, and thus can more precisely explain how input features affect the output.

[0059] In order to verify the effectiveness of the method of this embodiment, an experiment was conducted on the dynamic pricing model of reinforcement learning based on Stackelberg game using a real data set, where the real data set is derived from existing public data and the consent of the information owner is obtained.

[0060] (1) Pricing benefits, Among them, KDQN represents the reinforcement learning dynamic pricing model based on Stackelberg game proposed in this invention, and the other algorithms are comparative algorithms, namely DQN: an algorithm that combines deep learning and reinforcement learning, using neural networks to approximate the Q-value function, and is used to solve decision-making problems in high-dimensional state spaces. RBFNN: a feedforward neural network, usually used for tasks such as function approximation, classification, and time series prediction. It uses radial basis function as the activation function and constructs a nonlinear mapping by calculating the distance between the input and the center point. TAPRIM: a task allocation and pricing incentive mechanism for edge-assisted crowd intelligence perception scenarios, which coordinates and optimizes the benefits of each participant in the system through a three-stage Stackelberg game model, while ensuring the rationality of task allocation and data quality.

[0061] like Figure 5 As shown in the figure, as the number of tasks increases, the performance of all models (KDQN, DQN, RBFNN, TAPRIM) in terms of pricing benefits shows an upward trend. However, the benefits of the KDQN model are significantly higher than those of other models, especially when the number of tasks is large, its benefits perform best. In contrast, the performance of the TAPRIM model is slightly worse than that of KDQN, but still better than DQN and RBFNN, showing good competitiveness.

[0062] Specifically, the KDQN model better approximates the profit maximization strategy through deep reinforcement learning, so its pricing profit is always ahead under different numbers of tasks.

[0063] The present invention conducts a large number of experiments on three real data sets to verify the performance of the proposed reinforcement learning dynamic pricing model based on Stackelberg game. The results show that the reinforcement learning dynamic pricing model based on Stackelberg game outperforms other baselines.

[0064] Example 2 This embodiment provides a mobile crowdsourcing strategy optimization system based on multi-objective optimization, including: The data acquisition module is configured to: use the crowdsourcing platform to publish task information, including publishing task types, maximum budgets, and equipment requirements; The push module is configured to: push information according to the publishing location of the task information, including extracting the geographical location where the task information is published, and push the task according to a preset broadcast range; The attribute module is configured to: calculate the worker information attributes based on the published task information and upload the attributes to the crowdsourcing platform, wherein the calculated worker information includes calculating the distance attribute, reputation attribute, minimum income and capability value attribute of the task; The feature extraction module is configured to: perform task-worker matching according to worker information attributes and task information, including performing task-worker matching using an EH-STLS algorithm, and generating an optimal task allocation information table; The model module is configured to: input the optimal task allocation information table into the KDQN dynamic pricing model, perform dynamic pricing operation through the Stackelberg game, and obtain an initial pricing model; The transformation module is configured to: use the initial pricing model to perform real-time dynamic pricing based on the current task and the actual situation of the worker, including combining the number of tasks completed by the worker on the day, the current reputation value, and the encryption agency's current reputation and current score.

[0065] A computer-readable storage medium stores a plurality of instructions, wherein the instructions are suitable for being loaded and executed by a processor of a terminal device, for a mobile crowdsourcing strategy optimization method based on multi-objective optimization.

[0066] A terminal device includes a processor and a computer-readable storage medium, wherein the processor is used to implement various instructions; the computer-readable storage medium is used to store multiple instructions, wherein the instructions are suitable for being loaded by the processor and executing the mobile crowdsourcing strategy optimization method based on multi-objective optimization.

[0067] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A mobile crowdsourcing strategy optimization method based on multi-objective optimization, characterized in that: include: Use crowdsourcing platforms to publish task information, including task type, maximum budget, and equipment requirements; Pushing information based on the location where the task information is published, including extracting the geographical location where the task information is published, and pushing the task based on a preset broadcast range; Calculate the worker information attributes based on the published task information and upload them to the crowdsourcing platform, where the calculated worker information includes the distance attribute, reputation attribute, minimum income and capability value attribute of the calculation task; Perform task-worker matching based on worker information attributes and task information, including using the EH-STLS algorithm to perform task-worker matching and generate an optimal task allocation information table; The optimal task allocation information table is input into the KDQN dynamic pricing model, and dynamic pricing is performed through the Stackelberg game to obtain the initial pricing model; The initial pricing model is used to conduct real-time dynamic pricing based on the current tasks and the actual situation of the workers, including combining the number of tasks completed by the workers on that day, the current reputation value, the current reputation of the encryption agency, and the current rating.

2. According to the mobile crowdsourcing strategy optimization method based on multi-objective optimization in claim 1, it is characterized in that: The worker information attributes are calculated based on the published task information and uploaded to the crowdsourcing platform, including calculating the distance attribute according to the geographic information in the task information and the current geographic information of the worker, and normalizing the calculated distance attribute. The distance attribute calculation formula is: ， in, is the radius of the Earth, is the dimensional difference between the worker's location and the task's posting location, is the dimension of the worker’s current geographical information, is the dimension of the task publishing location, is the longitude difference between the worker's location and the task's posting location.

3. The mobile crowdsourcing strategy optimization method based on multi-objective optimization according to claim 1 is characterized in that: The use of the EH-STLS algorithm for task-worker matching includes preliminary screening of worker information attributes of all uploaded information according to task information and worker information attributes to obtain a list of worker candidates, and generating an initial solution set based on the list of worker candidates, and using the EH-STLS algorithm to perform a comprehensive evaluation in the initial solution set, so that each task is matched with only one worker at this stage, thereby generating a preliminary global optimal solution set.

4. The mobile crowdsourcing strategy optimization method based on multi-objective optimization according to claim 1 is characterized in that: The use of the EH-STLS algorithm for task-worker matching also includes performing dynamic crossover mutation iteration based on the initial solution set and satisfying requirements and constraints, and after the iteration is completed, further searching for high-quality Pareto solutions through variable domain local search to generate an optimal task allocation information table.

5. The mobile crowdsourcing strategy optimization method based on multi-objective optimization according to claim 1 is characterized in that: The method of inputting the optimal task allocation information table into the KDQN dynamic pricing model includes parsing and preprocessing the data in the task allocation information table, constructing a task-worker allocation matrix, extracting the task features and detailed attributes of the workers in the task allocation information table, and normalizing the extracted numerical data.

6. The mobile crowdsourcing strategy optimization method based on multi-objective optimization according to claim 1 is characterized in that: The dynamic pricing operation is performed through the Stackelberg game to obtain the initial pricing model, including using the reverse induction method to solve the optimal strategy of followers, solving the optimal work effort and pricing strategy of workers and encryption agencies respectively, and then solving the optimal pricing strategy of the platform, and then determining the Stackelberg equilibrium solution, building the initial pricing model based on the equilibrium solution, and obtaining the model function by fitting the relationship between the pricing strategy and the task requirements and budget.

7. The mobile crowdsourcing strategy optimization method based on multi-objective optimization according to claim 1 is characterized in that: The initial pricing model is used to perform real-time dynamic pricing based on the current tasks and the actual situation of the workers, including obtaining the number of tasks completed on the day and the current reputation value in real time for the workers, and integrating them with the worker ability information in the initial pricing model into a real-time state vector for the workers. For tasks, the task type, maximum budget and equipment requirements in the initial pricing model are combined to form a real-time state vector for the task, and finally summarized into a complete real-time state vector for pricing calculation.

8. A mobile crowdsourcing strategy optimization system based on multi-objective optimization, characterized in that: include: The data acquisition module is configured to: utilize the crowdsourcing platform to publish task information, including publishing task type, maximum budget and equipment requirements; The push module is configured to: push information according to the publishing location of the task information, including extracting the geographical location where the task information is published, and push the task according to a preset broadcast range; The attribute module is configured to: calculate the worker information attributes based on the published task information and upload the attributes to the crowdsourcing platform, wherein the calculated worker information includes calculating the distance attribute, reputation attribute, minimum income and capability value attribute of the task; The feature extraction module is configured to: perform task-worker matching according to worker information attributes and task information, including performing task-worker matching using an EH-STLS algorithm, and generating an optimal task allocation information table; The model module is configured to: input the optimal task allocation information table into the KDQN dynamic pricing model, perform dynamic pricing operation through the Stackelberg game, and obtain an initial pricing model; The transformation module is configured to: use the initial pricing model to perform real-time dynamic pricing based on the current task and the actual situation of the worker, including combining the number of tasks completed by the worker on the day, the current reputation value, and the encryption agency's current reputation and current score.

9. A computer-readable storage medium storing a plurality of instructions, characterized in that: The instructions are suitable for being loaded by a processor of a terminal device and executing a mobile crowdsourcing strategy optimization method based on multi-objective optimization as described in claim 1.

10. A terminal device, comprising a processor and a computer-readable storage medium, wherein the processor is used to implement each instruction; and the computer-readable storage medium is used to store multiple instructions, characterized in that: The instructions are suitable for being loaded by a processor and executing the mobile crowdsourcing strategy optimization method based on multi-objective optimization as described in claim 1.

Images