摘要
移动众包平台是近几年兴起的一种自助式服务模式,用户通过自主选择任务,完成后获得相应的酬金,在确定每个新的任务定价上,传统的方式是将各个影响因素与定价之间进行多元回归分析,根据回归方程得出新的任务定价。但是这种方法存在拟合程度低,并不能很好地反映定价与各个影响因素的关系等问题,针对此问题,采用BP神经网络算法,运用MATLAB神经网络工具箱,将已完成的任务的定价视为合理的定价,对已完成的任务的定价规律进行学习,最后利用训练好的网络对于其他任务重新定价,实现对合理定价规律的学习和整体定价方案的优化。
The mobile crowdsourcing platform is a self-service service model which has emerged in recent years.Users select tasks by themselves and receive corresponding emoluments after completion.In determining the pricing of each new task,the traditional method is to use multiple regression to analyze the various influential factors and pricing,then predict the new task price based on the regression equation.However,this method has a low degree of fitting,and does not respond well to he relationship between various factors and price.To solve this problem,uses BP neural network algorithm and uses MATLAB neural network toolbox to complete tasks that have already been completed.These tasks are regarded as reasonable items that represent the reasonable pricing rules.Finally,uses the trained network to reprice other tasks and realizes learning of reasonable pricing rules and optimization of overall pricing plans.
作者
蒋师贤
杨亮
JIANG Shi-xian;YANG Liang(School of Information Engineering,Tianjin University of Commerce,Tianjin 300134)
出处
《现代计算机》
2018年第13期21-23,共3页
Modern Computer
基金
"大学生创新创业训练计划"项目(大创)(No.201610069004)
