基于Schulze社会选择理论的在线服务评价研究与实现

发布时间：2018-05-13 11:13

本文选题：在线服务 + 社会选择理论　；参考：《昆明理工大学》2017年硕士论文

【摘要】：在线服务作为一种利用互联网技术向用户提供线上服务的方式,已在电子商务、电子政务、e-Learning、e-Science等领域迅速普及并得到广泛应用。随着在线服务的兴起,网络上出现了大量具有相似功能的服务。因此,用户需要对使用哪些服务以满足自身需求进行决策。然而,在互联网环境下,用户从众多具有同样功能的服务中进行选择决策存在多方面的困难,使用户不可避免地面临不能获取真正需要的服务的风险:一方面,由于在线服务数量庞大,用户不可能与每一个服务都有交易经验,需要花费大量时间比较每种服务的优劣。另一方面,由于网络的虚拟性使用户不能对在线服务有直接的了解和接触,而商家为了提高自身的信誉,可能向用户提供不真实的产品信息。因此,用户需要借助第三方观点为基础形成的评价对在线服务的优劣进行判断就显得尤为重要。目前在线电子商务评价方法主要有累加法、平均法、概率法、模糊法、流程法等,但这些方法没有考虑到用户评分标准不一致导致不同用户对同一服务评分不可比较的问题。为在不同用户对同一服务评分不可比较时提高在线服务评价的客观性,本文提出了一种基于Schulze社会选择理论的在线服务可比较评价方法。首先基于用户之间的相似性对不完全评分数据进行填充;然后根据填充得到的用户-服务评分矩阵计算用户对在线服务的偏好关系;最后利用群决策方法中的Schulze社会选择函数,通过构建加权有向图并计算最强路径得到服务评价。本文选用Schulze社会选择函数进行在线服务的评价,并对Schulze函数用于在线服务评价的适用性和有效性进行了理论分析和实验验证。理论分析表明本文提出的方法具备有效的评价方法应当具有的孔多塞性、单调性、非独裁性和操纵复杂性。实验结果也表明Schulze函数适用于在线服务的评价。最后,根据本文提出的方法,设计实现了在线服务评价原型系统。
[Abstract]:As a way to provide online services to users using Internet technology, online services have been widely used in the fields of electronic commerce, e-government, e-learning and e-Science. With the rise of online services, a large number of similar services appear on the network. Therefore, users need to make decisions about which services to use to meet their own needs. However, in the Internet environment, there are many difficulties for users to choose and make decisions from a large number of services with the same functions, which make users inevitably face the risk of not obtaining the services they really need: on the one hand, Because of the large number of online services, it is impossible for users to have trading experience with each service, and it takes a lot of time to compare the merits and demerits of each service. On the other hand, because of the virtual nature of the network, users can not directly understand and contact with online services. In order to improve their own credibility, merchants may provide users with untrue product information. Therefore, it is very important for users to judge the merits and demerits of online services by means of the evaluation based on third party views. At present, there are several online e-commerce evaluation methods, such as cumulative method, average method, probability method, fuzzy method, flow method and so on. However, these methods do not take into account the problem that different users can not compare the scores of the same service due to the inconsistency of the user rating criteria. In order to improve the objectivity of online service evaluation when different users' scores on the same service are not comparable, this paper presents an online service comparable evaluation method based on Schulze's social choice theory. Firstly, the incomplete scoring data is filled based on the similarity between users; then the user preference relationship to online services is calculated according to the filled user-service score matrix; finally, the Schulze social selection function is used in group decision making. The service evaluation is obtained by constructing weighted digraph and calculating the strongest path. In this paper, Schulze social selection function is used to evaluate online service, and the applicability and effectiveness of Schulze function in online service evaluation are analyzed theoretically and experimentally. The theoretical analysis shows that the method proposed in this paper has the characteristics of Compressionality, monotonicity, non-autocracy and complexity of manipulation. The experimental results also show that the Schulze function is suitable for online service evaluation. Finally, according to the method proposed in this paper, an online service evaluation prototype system is designed and implemented.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP311.52

【参考文献】