基于神经网络和SNA的C2C电子商务信誉欺诈识别研究*

当前,互联网中信誉欺诈行为严重影响了C2C电子商务的发展。如何有效识别互联网上的信誉欺诈商户是当前的研究热点,文中阐述了将基于交易历史的社会网络分析用于构建C2C电子商务信誉欺诈识别指标体系的原理及过程,通过使用LVQ神经网络对雅虎奇摩拍卖网采集的大量用户交易数据进行分类,有效识别出控制多个虚假账户以达到信用累计目的的信誉欺诈商户,对维护C2C电子商务交易的稳定性具有重要意义。     

基于群组的C2C电子商务信任模型研究

国内的电子商务平台大多采用eBay信任模型,对买家反馈评分进行简单累加得到卖家信誉值,未区分买家反馈评分的合理性及其参考价值的重要程度。为此,提出一种基于群组的C2C电子商务信任模型,通过计算买家与卖家的熟悉度,确定买家的可信度,综合考虑反馈评分、交易次数、交易价格、交易时间、以往买家的可信度对信誉的影响,构建电子商务信任模型。模拟实验结果证明,相比Sporas模型,该模型能为买家提供更全面、更准确的卖家信誉信息。     

基于社会网络与信誉的C2C信任评价模型

将概率论和社会网络理论应用到信任传递以及多信任路径建模中,针对C2C交易平台实际的商业网络特征,提出基于社会网络与信誉的C2C信任评价模型。引入相似度计算建立簇与簇之间的关联,以解决C2C交易网络存在的买卖分工明确、稀疏性、聚簇问题。对淘宝网C2C交易平台数据的分析表明,该模型对淘宝网现有信誉机制具有较好的改进和补充作用,更符合现实生活中人与人之间的信任和推荐关系。     

C2C电子商务信任评价模型设计与实现

目前C2C电子商务环境下使用的信任评价模型存在诸多问题,提出基于交易成功率的C2C电子商务信任评价模型,并从模型框架、信任值计算、信任评价模型算法、用户信任值存储、仿真实验等多个方面对模型进行阐述。通过仿真实验,从评价准确率和交易成功率两点就本模型与淘宝的信任评价模型进行比较。     

一种抗欺诈的C2C卖方信誉计算模型研究

针对C2C信誉模型中小额商品信誉炒作、信誉共谋、信誉诋毁等问题,引入交易价格、反馈可信度、共谋因子等参数,提出一种买方视角下抗欺诈的卖方成员信誉计算模型(C2CRep)。实验中通过收集网络交易数据,定义可疑欺诈的基本特征对数据进行抽取,并设定信誉计算误差(RCE)指标检验由欺诈行为带来的信誉值在社区信誉所占比例来检验模型的应用效果。结果表明,C2CRep在3类不同比例的欺诈行为中,RCE明显低于SPORAS与淘宝信誉模型,且RCE值在3类实验中都低于15%,抗欺诈性强。     

基于信任网络的C2C电子商务信任算法

C2C电子商务交易具有匿名性、随机性、动态性的特点,交易双方仅通过虚拟网络交换信息,缺乏基本的信任基础,交易存在较大的风险.构造科学的信任计算模型、客观度量卖家的可信度、辅助买家(消费者)做出正确的购买决策,是降低交易风险的有效手段之一.为此,从买家的角度出发,详细讨论了信任网络的基本概念及其相关属性,并以信任的时间敏感性、不对称性、可传递性和可选择性为基础,建立了C2C电子商务环境下的动态信任算法(C2C dynamic trust algorithm,简称CDTA).该算法首先通过买家自身的交易经验计算买家对卖家的直接信任度,然后计算来自信任网络中买家的朋友对卖家的推荐信任度,最后通过信任调节因子集成直接信任度和推荐信任度来获得买家对卖家的信任度.仿真实验分析结果表明:一方面,该算法考虑了交易的多属性及其相关性,信任评价的粒度更加细化,使得信任计算的结果更加客观;另一方面,评价相似度可以很好地筛选出符合买家“个性”的推荐节点,使推荐信任度更准确,可以进一步抑制恶意节点对信任算法的影响.     

C2C电子商务中基于声誉的信任评估模型

C2C模式电子交易平台以其便利、自由和高效性,而得到迅速发展,但由于其中信任评价机制的不完善,使得对交易参与者的监管难度增大,交易中仍存在较大的风险。针对目前信任评价机制的不完善,提出了基于声誉的信任机制,通过计算交易参与方的信任度以达到识别不诚实的个体的目的。通过进一步研究和分析影响评价的因素,使用时间窗口和交易窗口机制对参与计算的评价进行选择,并综合考虑评价者的自身信任度。对模型的有效性和效率都进行了理论分析和实验验证。结果表明,提出的模型能够有效地应用于C2C电子商务系统中。     

参与式感知系统中用户累积行为信誉计算

参与式感知系统中,由于感知数据质量可能受参与者影响,提出了基于用户累积行为的信誉计算模型以帮助选择可信赖用户.针对感知环境中用户群体的广泛性及核心用户的不确定性,该模型采用OPTICS聚类算法定义用户场景并划分行为数据集,建立用户累积行为信誉计算模型,同时引入时间戳标记信息抛弃部分旧行为以更新用户信誉.实验表明,该信誉模型能够结合新旧行为较好地计算并调整用户信誉,在感知环境用户信誉评价中具有良好的应用前景.     

C2C信用评价体系的风险计算

目前,C2C电子商务网站纷纷建立了较合理的评价体系,但在交易中仍然存在风险问题。本文通过对C2C模式交易过程的分析,提出了一种根据历史交易的情况,综合考虑卖家的信用值、买家的信用等级以及商品价格等因素的信用风险计算方法。通过模拟实验表明,对目前网上购物的信用炒作起到了一定的防御作用,有助于交易者做出正确的交易决策,为电子商务网站提供了更有效的信用评价信息。     

物流货运交易网站信用评价模型研究

信用问题是物流行业发展所面临的重要问题,也是C2C模式网上货运交易的主要瓶颈之一。首先对目前大多数货运交易网站所采用的信用评价模型进行了研究,指出现有模型所存在的三个主要问题：没有考虑用户自身信用水平的影响、缺少对交易风险大小的区分、对交易欺骗惩罚不足。然后,针对已有模型的不足,提出了改进的交易信用评价模型并设计了评分算法。最后,通过在模拟环境中与原模型进行对比,验证了改进模型和算法的有效性。     

Parameter learning of personalized trust models in broker-based distributed trust management

Distributed trust management addresses the challenges of eliciting, evaluating and propagating trust for service providers on the distributed network. By delegating trust management to brokers, individual users can share their feedbacks for services without the overhead of maintaining their own ratings. This research proposes a two-tier trust hierarchy, in which a user relies on her broker to provide reputation rating about any service provider, while brokers leverage their connected partners in aggregating the reputation of unfamiliar service providers. Each broker collects feedbacks from its users on past transactions. To accommodate individual differences, personalized trust is modeled with a Bayesian network. Training strategies such as the expectation maximization (EM) algorithm can be deployed to estimate both server reputation and user bias. This paper presents the design and implementation of a distributed trust simulator, which supports experiments under different configurations. In addition, we have conducted experiments to show the following. 1) Personal rating error converges to below 5% consistently within 10,000 transactions regardless of the training strategy or bias distribution. 2) The choice of trust model has a significant impact on the performance of reputation prediction. 3) The two-tier trust framework scales well to distributed environments. In summary, parameter learning of trust models in the broker-based framework enables both aggregation of feedbacks and personalized reputation prediction.

Securing decentralized reputation management using TrustGuard

Reputation systems have been popular in estimating the trustworthiness and predicting the future behavior of nodes in a large-scale distributed system where nodes may transact with one another without prior knowledge or experience. One of the fundamental challenges in distributed reputation management is to understand vulnerabilities and develop mechanisms that can minimize the potential damages to a system by malicious nodes. In this paper, we identify three vulnerabilities that are detrimental to decentralized reputation management and propose TrustGuard—a safeguard framework for providing a highly dependable and yet efficient reputation system. First, we provide a dependable trust model and a set of formal methods to handle strategic malicious nodes that continuously change their behavior to gain unfair advantages in the system. Second, a transaction-based reputation system must cope with the vulnerability that malicious nodes may misuse the system by flooding feedbacks with fake transactions. Third, but not the least, we identify the importance of filtering out dishonest feedbacks when computing reputation-based trust of a node, including the feedbacks filed by malicious nodes through collusion. Our experiments show that, comparing with existing reputation systems, our framework is highly dependable and effective in countering malicious nodes regarding strategic oscillating behavior, flooding malevolent feedbacks with fake transactions, and dishonest feedbacks.     

The dynamics of trust in B2C e-commerce: a research model and agenda

Trust is a key factor that determines the success of Business to Consumer (B2C) e-commerce transactions. Research has identified several critical factors that influence trust. These have been incorporated into the design of e-commerce web portals; however a vast majority of potential users are still wary of online business transactions. This points to the need for the development of more sophisticated and comprehensive models that can provide additional insights into the true role played by trust in the adoption of e-commerce portals. Here we do so by focusing on trust as it pertains to B2C e-commerce transactions. Pertinent issues addressed are: (1) Is trust the critical factor that makes all the difference between users accepting or rejecting a B2C portal? (2) Are there factors that could play a more critical role in user acceptance/rejection of such portals? (3) If so, what is the relevance of trust relative to these factors? The research model and agenda introduced here enable researchers to view trust in its proper perspective as part of the bigger picture of technology acceptance.     

A Fine-Grained Reputation System for Reliable Service Selection in Peer-to-Peer Networks

Distributed peer-to-peer (P2P) applications have been gaining momentum recently. In such applications, all participants are equal peers simultaneously functioning as both clients and servers to each other. A fundamental problem is, therefore, how to select reliable servers from a vast candidate pool. To answer this important open question, we present a novel reputation system built upon the multivariate Bayesian inference theory. Our system offers a theoretically sound basis for clients to predict the reliability of candidate servers based on self-experiences and feedbacks from peers. In our system, a fine-grained quality of service (QoS) differentiation method is designed to satisfy the diverse QoS needs of individual nodes. Our reputation system is also application-independent and can simultaneously serve unlimited P2P applications of different type. Moreover, it is semidistributed in the sense that all application-related QoS information is stored across system users either in a random fashion or through a distributed hash table (DHT). In addition, we propose to leverage credits and social awareness as reliable means of seeking honest feedbacks. Furthermore, our reputation system well protects the privacy of users offering feedbacks and is secure against various attacks such as defaming, flattering, and the Sybil attack. We confirm the effectiveness and efficiency of the proposed system by extensive simulation results.     

一种时域上的P2P信任模型

传统的P2P信任是根据节点间交易成功和失败的次数来合成的,该值是个累积量,而实际的信任值是时变量。提出了一种新的P2P时域信任模型,对于局部信任,通过对每个时间段合成一个信任,然后根据时间段的新近赋予不同的加权合成局部信任;对于全局信任来说,随着时间的推移,发起节点会越来越重视自身对目标节点的信任评价,而其它节点的推荐值会得以削弱。该模型刻画了信任的动态性,能够有效地防止不良节点的信任短期积聚以及依靠其它节点共谋等恶意行为。     

A reputation system for e-marketplaces based on pairwise comparison

Implementing a reputation system is an effective strategy to facilitate trust and security in an online environment. In addition to that, reputation systems can help online customers through decision-making process. However, in real-world situations, these systems have to deal with plenty of problems and challenges. This paper aims to solve four problems that are common to reputation systems in e-marketplaces, namely the subjectivity of ratings, inequality of transactions, multi-context reputation and dynamic behavior of users. The proposed model starts with the pairwise comparison, which is a powerful tool for removing bias from ratings. Then, we extend the concept of pairwise comparison to contests between users. A pairwise comparison has only a winner and a loser, but we can associate a score differential with a pairwise comparison when we consider it as a match. This score differential is adjusted in a way that three other problems can be solved. We implemented our model in a multi-agent simulation in which real-world data were also incorporated. We compared our model with some of previous reputation systems. Experiments show that our model outperforms previous ones when faced with real-world challenges.     

面向价值的O2O电子商务信任预测模型

电子商务市场所面临的信用欺诈制约了O2O电子商务的快速发展。针对电子商务交易各方利益不平衡的问题,设计了一种面向价值的O2O电子商务信任预测模型。在该模型中,综合考虑了交易评价、交易时间、交易金额、交易次数、已有声誉值、评价的可信度等多种影响因素,利用置信因子对局部声誉和全局声誉进行整合,预测线下商家和消费者的信任度。实验证明,该信任模型能够有效计算线下商家和消费者的动态信任度,并对不诚信交易行为具有较强的防御能力,可实时更新O2O电子商务交易各方的信任度。