中文摘要：

O2O电子商务近年来发展迅猛,然而部分O2O服务提供商存在不能诚信经营的问题.面对浩如烟海的O2O服务提供商,用户难以做出对自己有利的选择.针对这一问题,提出一种基于声誉计算的可信O2O服务提供商推荐方法.该方法根据用户服务评价矩阵计算用户对服务的加权评价,产生用户对商家的综合评价,进而得到用户商家综合评价矩阵.结合商家声誉,计算用户之间的O2O服务提供商偏好相似性,然后进行推荐.与现有的推荐方法相比,所提出的推荐方法能更有效地对O2O服务提供商的服务质量和经营是否诚信等问题进行过滤,同时考虑用户的偏好属性,具有更高的推荐准确度和更强的抗欺诈能力.

英文摘要：

As online to offiine （O2O） e-commerce develops quickly in recent years, many problems have occurred. Some service providers are poor, while some providers are dishonest. It is difficult for users to choose an honest O2O service provider who can provide suitable service. To solve the problem, we propose a method for trustworthiness recommendation of suitable O2O service providers to users based on the providers＇ reputation and users＇ similarity. A user-service rating matrix is used to compute the users＇ comprehensive ratings on O2O service providers, and a comprehensive user-service provider rating matrix generated. The recommendation method combines the matrix with reputations of the O2O service providers to compute similarities of different users. Suitable O2O service providers are then recommended to users. Simulation and experimental results demonstrate that the proposed method has better recommendation accuracy as compared to other traditional methods, as well as some state-of-the-art methods. It performs well in resisting malicious attacks.