A model of new workers' accurate acceptance of tasks using capable sensing

Crowdsourcing has been one of focuses in academic and industrial communities along with rapid development and wide-spread applications of Internet. However, the lack of a new worker's capacity of accepting tasks seriously affects his/her income obtained by fulfilling tasks issued by requesters,...
Ausführliche Beschreibung

Crowdsourcing has been one of focuses in academic and industrial communities along with rapid development and wide-spread applications of Internet. However, the lack of a new worker's capacity of accepting tasks seriously affects his/her income obtained by fulfilling tasks issued by requesters, which reduces his/her enthusiasm for participation in crowdsourcing. We propose a method of solving the problem of accurately accepting tasks for a new worker in this paper. To fulfill this task, we firstly formulate the problem as a constraint optimization problem with an unknown parameter which shows the time consumption in fulfilling a task by a new worker. Then, we estimate the time consumption using information about similar tasks and workers in the crowdsourcing platform. Finally, we generate a strategy of accepting tasks by solving the optimization problem using a genetic algorithm, with the purpose of maximizing the income of the new work. We evaluate the effectiveness of the proposed strategy based on data in Taskcn, a representative commercial crowdsourcing platform in China, by comparing the results with a number of workers' actual earning. The experimental results demonstrate the accuracy of the new worker's capacity of accepting tasks, which is beneficial for generating a strategy to improve his/her income. Ausführliche Beschreibung