The traditional task allocation method mainly depends on subjective judgment and personal experience, which will directly lead to low utilization efficiency of human resources. Considering the shortage in existing task allocation method including the complexity of human resource and difference in ability, a new type of personalized task allocation method is proposed in "Crowdsourcing", which not only consider the outsouree's comprehensive technical ability, but also take into account his personality attributes, and online reputation. So that the task allocation process is more targeted. This method is applied to "JuHuiNet", and experiments show that this method can provide an effective task allocation decision support for the Outsourcer.