移动商务下基于位置和偏好模型的服务推荐

发布时间：2018-04-04 13:11

本文选题：移动商务　切入点：位置　出处：《海南大学》2016年硕士论文

【摘要】：随着移动互联网的普及,人民生活进一步的“互联网化”。为了满足用户随时随地的移动需求,特别是在移动医疗、移动金融等新兴领域,移动应用程序需要在同一时间多个维度向用户提供服务。在互联网中移动互联网业务大致可分为：手机在线支付、网上购物、网上银行和旅游预订等移动应用,它们同比增长率都在40%以上。其中移动互联网用户是主要推动力,这为移动商务的进一步发展提供了良好的基础。与传统的电子商务相比较而言,移动商务的服务对象具有高度的服务性和流动性。因此,为了适应移动商务的新特性,必须在满足用户购买需求的基础上,对用户进行服务定制。针对这一现状,本文对移动商务下的服务推荐系统进行了深入的研究。本文的主要工作如下：第一,阐述了在移动商务环境下开展服务推荐的意义,从推荐算法动态适应用户兴趣变化的角度总结了国内外研究现状。同时根据移动商务用户特点,从位置和浏览路径两方面对移动商务下的服务推荐算法进行深入分析。第二,利用移动商务环境下,系统可以实时获取用户地理位置的优势,在传统推荐算法的基础上引入距离变量。根据用户浏览商品的不同,实时调整距离变量,进而向用户推荐最合适的商品。该方法解决了由于外部信息改变带来的用户短期兴趣变化的问题。第三,在引入位置解决用户短期兴趣的基础上,考虑到移动商务环境下,用户数据集的稀疏性、冷启动问题,采用蚁群算法依据用户的浏览路径,提高算法的推荐精度。最后,通过仿真实验,验证在移动商务下引入距离变量和蚁群算法的可行性。实验表明,引入距离变量以后推荐系统可以根据用户具体的距离敏感度来为用户提供更加适合当前环境的服务推荐,基于蚁群的用户浏览路径算法可以有效的提高推荐的精度。
[Abstract]:With the popularity of the mobile Internet, people's life further "Internet."In order to meet the mobile needs of users, especially in the emerging areas of mobile medicine, mobile finance and so on, mobile applications need to provide services to users in multiple dimensions at the same time.Mobile Internet services in the Internet can be broadly divided into mobile applications such as mobile online payment, online shopping, online banking and travel booking, all of which are growing at more than 40 percent year-on-year.Mobile Internet users are the main driving force, which provides a good basis for the further development of mobile commerce.Compared with the traditional e-commerce, mobile commerce has a high degree of service and mobility.Therefore, in order to adapt to the new features of mobile commerce, it is necessary to customize the service to users on the basis of meeting the purchase needs of users.In view of this present situation, this paper carries on the thorough research to the service recommendation system under the mobile commerce.The main work of this paper is as follows: first, the significance of service recommendation in the mobile commerce environment is expounded, and the research status at home and abroad is summarized from the point of view of dynamic adaptation of recommendation algorithm to the change of user interest.At the same time, according to the characteristics of mobile commerce users, the service recommendation algorithm under mobile commerce is analyzed from two aspects: location and browsing path.Secondly, under the mobile commerce environment, the system can obtain the advantage of users' geographical location in real time, and introduce the distance variable on the basis of the traditional recommendation algorithm.The distance variable is adjusted in real time to recommend the most suitable item according to the different items viewed by the user.This method solves the problem of the change of user's short-term interest caused by the change of external information.Thirdly, on the basis of introducing location to solve the short-term interests of users, considering the sparse and cold start problem of user data set in the mobile commerce environment, ant colony algorithm is used to improve the recommendation accuracy of the algorithm according to the browsing path of users.Finally, the feasibility of introducing distance variables and ant colony algorithm in mobile commerce is verified by simulation experiments.The experimental results show that the recommendation system can provide users with more suitable service recommendation for the current environment according to the specific distance sensitivity after the introduction of distance variables. The user browsing path algorithm based on ant colony can effectively improve the accuracy of recommendation.
【学位授予单位】：海南大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP391.3;TP18

【参考文献】