基于信任模型的鲁棒众包数据分析方法

Robust Crowdsource Data Analysis Method Based on Trust Model

众包是互联网大发展趋势下衍生的一种非常流行的新型商业模式,企业将过去由员工执行的任务分配出去,以自由自愿的形式外包给非特定的(通常是大型的)大众志愿者来完成。但是,在这一过程中,有些志愿者并没有认真地完成任务,为了骗取佣金,使利益最大化,往往会提供虚假数据,造成任务结果数据虚实混杂,雇主无法得到高准确度的众包任务结果。针对这个问题,在对现有众包数据分析方法进行分析之后,提出了一种新的众包数据分析方法。在考虑工作者历史信誉度的基础上,对部分任务结果采用投票一致性规则来分析,将工作者历史信誉度和在本次任务中提交的结果数据与贝叶斯算法模型相结合,得出工作者在本次任务中提交结果数据的最终准确度信息。实验结果表明,使用该方法筛选出的工作者可靠性更高。

Crowdsourcing is a quie popular new business model derived from the great development trend of the Internet.Enterprises assign tasks previously performed by employees to non-specific(usually large)public volunteers in a free and voluntary manner.However,in this process,some volunteers did not complete the task seriously.In order to cheat the commission and maximize the benefits,they often provided false data,resulting in mixed actual and actual task result data,so the employers could not obtain high-accuracy crowdsourcing task results.To solve this problem,after analyzing the existing crowdsourcing data analysis methods,we propose a new crowdsourcing data analysis method.On the basis of considering the historical credibility of workers,we use voting consistency rules to analyze some task results.The historical credibility of workers and the result data submitted in this task are combined with Bayesian algorithm model to obtain the final accuracy information of the result data submitted by workers in this task.The experiment shows that the workers screened by this method have higher reliability.