Task allocation method based on genetic algorithm under crowdsourcing environment

Abstract

The invention relates to a task allocation method based on a genetic algorithm under the crowdsourcing environment. The method comprises the steps of receiving a submission task of a task submitter and a plurality of decomposed subtasks, receiving worker information participating task bidding provided by a crowdsourcing platform, constructing skill layering trees of participant skill relations, obtaining values capable of being provided by workers for sub-tasks, extracting keyword information, obtaining keyword bases of workers, obtaining the similarity among the workers, adopting a genetic algorithm for solving a task allocation scheme, and performing task allocation through the scheme obtained by the crowdsourcing platform. The defects that time and labor are wasted and the effect is poor are overcome. The most suitable team is allocated to one task, in this way, the final task delivery quality can be improved, and workers can better cooperate with workers with the consistent development experience as much as possible when completing the task.

Description

technical field [0001] The invention relates to a method for allocating subtasks and bidding participants of a relatively complex task on a crowdsourcing platform, in particular to a task allocation method based on a genetic algorithm in a crowdsourcing environment. Background technique [0002] In recent years, crowdsourcing has received extensive attention from both industry and academia. The concept of crowdsourcing (crowdsourcing) was proposed by Jeff Howe (Jeff Howe), a reporter from the American "Wired" magazine, in June 2006. Jeff Howe's definition of "crowdsourcing" is: "A company Or the free and voluntary outsourcing of work tasks performed by employees in the past to non-specific (and usually large) public networks. Crowdsourcing tasks are usually undertaken by individuals, but if it involves multiple people Collaboratively accomplished tasks may also appear in the form of individual productions that rely on open source." [0003] Today, crowdsourcing platforms h...

AI Technical Summary

Problems solved by technology

[0004] However, the task allocation scheme of the current crowdsourcing platform often requires workers to search and select the tasks they want to participate in, and then the task submitter or the platform manually assigns them, which is time-consuming and laborious, and often does not get better results.

Some task assignment methods under the only crowdsourcing platform usually only start from the perspective of tasks, only focus on the connection between tasks and workers, and do not take into account the cooperation affected by the similarity between workers. Relationships are also a factor in the quality of final task delivery

In addition, some current crowdsourcing task allocation strategies usually require a strict match between worker skills and the skills required for the task, which often leads to allocation failures due to workers without strict skill matching.

Images