联系我们

论文查询结果	返回搜索

论文编号：	1034
作者编号：	2120071912
上传时间：	2009/5/26 23:46:02
中文题目：	基于云模型的C2C卖家信任评价方
英文题目：	Research on Evalument Method o
指导老师：	侯文华
中文关键字：	C2C;信任;云模型;评价方法tl
英文关键字：	C2C;trust;cloud model;evalumen
中文摘要：	近年来，国内电子商务发展十分迅速，取得了可喜的成果，然而电子商务的却面临着缺乏消费者信任的发展障碍。在电子商务模式下，由于交易双方信息不对称程度的加大和网络欺诈行为的不断发生，使得消费者对网上交易所感知的风险大大增加，不信任感增大。实际上，信任问题己经成为我国电子商务发展的最大障碍。建立电子商务中消费者的信任成为有效降低消费者的感知风险，进而进行网上购买的关键。本文的目的在于研究C2C模式下的如何对卖家的信任等级进行判定的方法。 本文使用云模型理论和方法，在淘宝网评价体系基础上，定义了属性评价云、综合信任云和信任云三种概念。这三种概念是相互在对方的基础上建立起来的，是具有层次递进关系的。通过对卖家信任等级评价方法的研究，本文对原有的综合信任评价方法进行了改进，加入了熵权作为权重参与综合信任云的合成和多维云相似系数的计算。并在此基础上利用云相似系数划分了信任等级。本文将信任等级分成了三种好评、中评和差评。改进后的方法更加适用于本文所使用的评价体系进行信任等级的评定。根据卖家对应这三种等级所具有的综合信任云的数量，使用逆向云发生器生成卖家的信任云，并把该云的期望参数作为该卖家的信任度。但卖家的信任等级不可以单一只看信任度。为了对方法的可操作性和科学性进行验证，本文对方法进行了案例仿真。案例仿真的结果说明本文所研究的方法是科学有效的，并有较高的实用和商用价值。 本文的研究成果主要有对基于云的信任评价方法进行了多方面的改进，并将信任等级分成了具体的三个类别，即好评、中评和差评。最终评价的结果也是用云的形式来表述，同时作者对结果进行了分析和解释。最后通过对淘宝网一真实卖家的交易评级记录使用该评价方法，得到了该卖家的信任云和信任度，并将结果与淘宝的评价方法进行了比较，并做了相关解释，分析了各自的优劣。案例仿真的结果表明模型是科学有效地。文章的最后作者给出了未来该方法的发展方向和一些有待改进的地方。
英文摘要：	In recent years, e-business has developed rapidly, and achieved encouraging results. It facing a lack of consumer trust, barriers to its development. Consumers feel risk of online exchanges and the sense of distrust as well, because of the increasing degree of information asymmetry and ongoing network fraud between the two sides of transaction. In fact, the problem of trust in our country has become the biggest obstacle to the development of electronic commerce. The establishment of consumer trust in e-commerce is a key to reduce the risk that was realized by consumers, and then make them purchase online. The purpose of this thesis is to examine the method of how to determine the seller’s trust degree in C2C mode. In this paper, the definition of the attributes evaluation cloud, trust cloud and integrated trust cloud was based on the Taobao’s credit appreciation system using cloud model theory and methods. These three concepts are interrelated with each other progressively. Through the research of sellers’ trust evaluation methods, this thesis makes an improvement to the original integrated evaluation methods. Entropies are added as weight into the integrating of synthesis trust cloud and the calculation of cloud similarity coefficient. And the levels of trust are seperated on the basis of cloud similarity coefficient. There are three levels specific for the trust, positive comment, average comment and negative comment. This improved method is fit for the evaluation system used in this paper to calculate sellers trust level. Simulation results show that the case studied in this paper is scientific and effective, and has high utility and commercial value. In this thesis, the results of research are mainly about a number of improvements based on the trust cloud evaluation method, and another contribution is making the trust level divided into three specific categories. Final evaluation results will be used with the expression of cloud form, and then this result will be analyzed and explanted by the author. Finally, through rating a real Taobao seller’s trust level using this evaluation method with real business transaction record. As a result, we get that seller’s trust cloud and trust degree. Case simulation results show that the model is a scientific and effective manner. The future direction of this method for development and a number of areas for improvement is given by the author in the end of the thesis.
查看全文：	预览  下载(下载需要进行登录)