Research On Space Crowdsourcing Task Allocation Based On Social Network Trust

With the rapid development of science and technology,crowdsourcing,as an important way to collect data,has gradually entered the public view,which has been paid more and more attention by more scientists.this paper evaluates worker trust values from a social network perspective and applies them to spatial crowdsourcing task assignment.Consider improving the task allocation system from the perspective of individuals and groups.Its main tasks are:(1)Taking into account the complex social network relationship of workers themselves,a multi-constrained path selection model based on worker reputation,social intimacy and reputation MCOPS(Multi-Constrained Optimized Path Selection)is established to improve the traditional Dijkstra algorithm to solve the optimal trust path from task publisher to target worker,and to calculate the trust value of departure contractor to worker.(2)Combining social network with spatial crowdsourcing to assign tasks to individuals,considering the influence of worker trust value and task incentive on task assignment,a Trust-aware Task Allocation algorithm(TTA)is proposed,and the travel cost and task urgency of workers are analyzed based on different scenarios,Space-weighted Trust-aware Task Allocation algorithm(STTA)and Time-weighted Trust-aware Task Allocation algorithm(TTTA).By using the minimum cost maximum flow algorithm and the Twitter check-in data as the real data set,the feasibility and practicability of the system in the face of different task scenarios are verified by comparing the success rate of task assignment.(3)Complex tasks often require multiple workers to complete the task together.This paper further assigns tasks to the group on the basis of(1).The task is divided into two parts.First of all,fuzzy estimation of the TWTW of tolerable waiting time for workers.Secondly,we discuss the constraints of group generation,put forward the importance of group consensus(Team)factors,abandon the traditional greedy task allocation algorithm,design the task assignment model based on tree decomposition,eliminate the dependencies between tasks,and finally find the optimal task traversal is used to find the optimal task allocation,and the global planning is carried out dynamically.In summary,this paper calculates the trust value of workers by mining the information data in social networks,and designs different algorithms to assign spatial crowdsourcing tasks to individuals and groups respectively.Considering various factors and using real data sets,the comparison of experimental data verifies the effectiveness of the proposed algorithm.