电子商务交易过程中信任评价的关键问题研究

发布时间：2018-03-15 11:13

本文选题：电子商务信任　切入点：多层信任　出处：《北京交通大学》2017年硕士论文　论文类型：学位论文

【摘要】：电子商务以其跨空间、费用低廉、机动灵活等优点,快速成为经济增长的新方式,然而在其快速发展的过程中,电子商务信任危机严重阻碍了电子商务的健康发展。剖析信任危机产生的原因,不合理的电子商务信任评价机制成为关键。无论是网站信任评价还是商家信任评价,现有信任评价机制都存在一定的缺失或漏洞,具体表现为不诚信网站多次威胁消费者财产安全、商家欺诈成本低、商家信任值的参考性较差等。因此,为帮助消费者建立电子商务信任,构建合理的电子商务信任评价模型刻不容缓。本文从电子商务交易过程出发,就电子商务信任评价中的两个关键问题——网站信任评价和商家信任评价,分别建立研究模型。网站信任评价是从网站的权威性、有用性、易用性、消费者服务、安全性和技术性六个方面综合衡量网站的可信程度,研究为消费者选择诚信网站或监管机构对网站信任排名提供理论依据;商家信任评价是在现有商家信任评价机制的基础上,综合交易满意度打分和交易属性等因素,构建更加科学合理的商家信任评价模型计算商家的信任值。本文的主要工作内容有:(1)基于已有信任和电子商务信任的相关研究,尝试性地提出了电子商务信任的概念,分析电子商务信任的特征与层次;从电子商务交易过程的角度,确定了电子商务信任评价的关键问题——网站信任评价和商家信任评价,并分析了两个关键问题的评价指标或影响因素。(2)网站信任评价的研究:首先,根据评价指标的量化方式获取并处理数据;其次,构建BP神经网络模型并通过平均影响值算法(MIV)筛选出8个关键评价指标;再次,比较确定MIV-BP神经网络模型作为网站信任评价模型;最后,将实例数据输入学习完成的MIV-BP神经网络模型获得实例网站的信任值与信任等级。(3)商家信任评价的研究以C2C电子商务为例:首先,基于商家信任评价的影响因素构建包含交易满意度、交易时间、交易金额、评价可信度、惩罚因子的商家信任评价模型;其次,采用C#编程实现模型与界面;最后,采集淘宝网卖家交易数据,针对恶意共谋评价、多次小额交易积累信任值、震荡交易等常见电子商务信任问题,与现有商家信任评价机制对比,验证模型的有效性与优越性。本文采用统计分析、机器学习、编程实现等研究方法,研究了电子商务信任评价的关键问题——网站信任评价和商家信任评价,研究成果对于完善电子商务信任研究、解决常见的电子商务信任问题提供了理论依据,具有较强的应用价值。
[Abstract]:The electronic commerce space, with its advantages of low cost, flexible, fast becoming a new way of economic growth, but in the process of rapid development, the electronic commerce trust crisis has seriously hindered the healthy development of e-commerce. Analyze the reasons of credit crisis, the mechanism of e-commerce trust evaluation unreasonable both become critical. Is the site of trust evaluation or business trust evaluation, existing trust evaluation mechanisms are missing or some loopholes, specific performance is not the integrity of the site repeatedly threatened consumers and property security, business fraud cost is low, the reference value of trust business is poor. Therefore, to help consumers establish e-commerce trust, build e-commerce trust evaluation model is reasonable urgent. This paper from the process of electronic commerce based on the two key problems in the evaluation of e-commerce trust -- trust evaluation website And the business of trust evaluation, respectively establish the research model. The evaluation from the authoritative website trust, website usefulness, ease of use, customer service, trusted degree of six safety and technical measure of comprehensive website, research for the consumer to choose the integrity of the site or regulatory agencies to provide a theoretical basis for ranking Website Trust business; trust evaluation is based on the existing business trust evaluation mechanism, comprehensive transaction satisfaction scoring and transaction attributes and other factors, calculation of businesses trust evaluation is more scientific and reasonable construction of the business model of the trust value. The main contents of this paper are: (1) research of trust and trust in electronic commerce based on the proposed the concept of trust in electronic commerce, analysis of the characteristics and level of trust in electronic commerce; from the electronic commerce transaction process point of view, the key problem to determine the electronic commerce trust evaluation- Website Trust Evaluation and merchants trust evaluation, and analyze the factors of the two key issues of the evaluation index or influence. (2) research on Website Trust Evaluation: first, according to the quantitative evaluation index of acquiring and processing data; secondly, construct the BP neural network model and algorithm through the influence of the average value (MIV) and selected 8 a key evaluation index; thirdly, comparing the MIV-BP neural network model as a web of trust evaluation model; finally, the MIV-BP neural network model instance data input learning completed for an example of a website trust value and trust level. (3) research on business trust evaluation based on C2C e-commerce as an example: firstly, factors affecting the business trust based on the evaluation of construction includes transaction satisfaction, transaction time, transaction amount, reliability evaluation, business trust evaluation and penalty factor model; secondly, using C# model and interface programming; finally, mining In taobao.com sellers transaction data against malicious collusion, many small transactions accumulated trust value, common e-commerce trust shocks transactions, with the existing business trust evaluation mechanism, effectiveness and superiority of the model. This paper uses statistical analysis, machine learning, programming and other research methods, key website trust the evaluation and assessment of the business trust E-commerce trust evaluation, research achievements will improve the e-commerce trust, provides a theoretical basis to solve the common problems of trust in e-commerce, has strong application value.

【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F724.6;F274

【参考文献】