Method for optimizing heterogeneous task execution sequence in crowd-sourcing system

Abstract

The invention discloses a method for optimizing the heterogeneous task execution sequence in a crowd-sourcing system. The method includes the steps of 1) a relation model f of task execution rates and loads of workers, 2) initializing the system, and inputting worker information, system time interval T and dead-line DL, 3) sequentially calculating the skill factor of each worker under different loads and the shortest execution time of each task in a task set in every time interval T, and adjusting the task execution sequence of each worker according to the sequence inequality principle, 4) and workers continuously executing the tasks in the task set in the time interval T of the system. Compared with the prior art, the method has the advantages that heterogeneity of tasks under the heterogeneous task environment is taken into consideration, which further optimizes the task execution time of workers; and that the task complete time can be shortened as much as possible through the greedy strategy based on the relation between manual task execution rate and the load of a worker.

Description

technical field [0001] The invention relates to the application of task allocation technology in a crowdsourcing system, in particular to a method for optimizing the execution sequence of heterogeneous tasks in the crowdsourcing system. Background technique [0002] With the popularization and application of the crowdsourcing system, for a batch of tasks, the system will select a reasonable matching method between the tasks and candidates according to the system goals and distribute the tasks sequentially. The task execution speed is fast, the cost is low, and the accuracy is high. It can complete some complex tasks that computers cannot complete. These characteristics in the crowdsourcing system have attracted more and more researchers to pay attention to the research on the crowdsourcing system. Among them, the task Allocation methods were especially the focus of the study. The crowdsourcing system mainly includes the task issuer, crowdsourcing platform, workers and tasks...

AI Technical Summary

Problems solved by technology

Considering the relationship between the rate at which workers perform tasks and the load, as well as the existence of heterogeneity among tasks, different task execution sequences will affect the worker's task completion time

The emergence of these new features makes the existing task allocation methods have limitations in the environment where the rate at which workers perform tasks is affected by the load and changes dynamically.

Images