基于粗糙集理论的第三方支付平台的效益问题

The Third-party Payment Platform Benefit Issues Based on Rough Set Theory

由于第三方支付平台交易者的效益问题存在着较多的不确定性,因此很少有定量化的研究成果。运用粗糙集理论,构建第三方支付平台交易的粗糙复杂网络,并对第三方支付平台效益进行研究,分别建立平台管理者效益提升的粗糙复杂网络信任模型、商家效益极大化的粗糙统计模型、顾客选择商品的粗糙决策模型;模型求解结果表明:平台管理者可以建立量化奖惩措施促进效益目标的实现;商家可以分析预测当广告费用为G0时的直接和间接销售量,从而得出是否做广告的决定;在价格适中且销售量适中、销售量高、商品价格低顾客评价适中且商家信用低、顾客评价高且商家信用高这四种情况下,顾客可以做出购买的决定。本研究在理论上不仅将粗糙集的理论用于不确定性复杂网络信任模型的研究,而且创立了基于时间序列分析的属性约简算法,从而解决了动态知识系统属性规则的提取问题。

There are many uncertainties on effectiveness of the third-party payment platform trading, so there is little quantitative research. Using rough set theory, we construct the third-party payment platform transaction＇s rough complex network, establish a rough complex networks trust model for platform managers to enhance efficiency, a benefit maximization rough statistical model for businesses and a rough decision model for customer choice goods. The results of the paper show that the platform manager may establish quantifiable incentives to promote the realization of the benefits goal through a trust model of rough complex network; business based on regression curve can analysis and predicts its direct and indirect sales when advertising costs are , to decide whether or not to advertise; customers would make a decision to buy only when the ＂moderate price and sales, or high sales, or low price and medium customer reviews and low business credits, or high business credit and high customer rating ＂. In this paper, not only in theory, rough set theory is used to study the uncertainty of complex network trust model, but also innovative given attribute reduction algorithm based on time series analysis to solve the attribute rules extraction problem in dynamic knowledge system.