基于粗糙集的电子商务推荐及可视化研究与实现

发布时间：2018-01-24 08:21

本文关键词： 电子商务推荐粗糙集理论可用性评估可用性测试　出处：《北京邮电大学》2010年硕士论文　论文类型：学位论文

【摘要】： 随着计算机的普及和互联网的发展,电子商务逐渐兴起并已经在当今社会零售业中占有较多的比重。围绕电子商务进行的各种研究方兴未艾,推荐系统便是其中一个重要的研究方向,对于电子商务的发展具有很大的现实意义。目前,几乎所有大型的电子商务网站,如Amazon,eBay,当当，，卓越,京东商城等,都不同程度的使用了各种形式的推荐系统。但当前对电子商务系统的研究主要侧重在各种推荐技术上面。对于电子商务网站推荐的可视化并没有太多的研究。本文首先研究了电子商务网站推荐技术的现状。对当前的推荐技术做了分析比较,同时提出一种基于粗糙集的用户购买倾向推荐机制,对推荐机制及算法做了研究和设计。通过对用户浏览行为记录的数据经过粗糙集理论中的约简,得出了用户行为与购买倾向之间的联系算法。其次本文通过对现有研究现状的分析,对当前主流电子商务网站做了以可用性测试技术为基础的分析和测试。得出了当前网站推荐可视化系统的不足之处。并设计了一个新的电子商务网站推荐系统。同时做了系统的高保真原型并进行了可用性评估测试,对新设计的推荐系统进行了验证。论文中做了大量的可用性测试,为系统的设计提供了从用户那里得到的许多建议和方法,最后的验证也说明了新系统的可用性更加出色,但论文中对于算法的验证没有进行过多的研究,也是后续需要进行改进的工作。
[Abstract]:With the popularity of computers and the development of the Internet, electronic commerce is gradually rising and has already occupied a large proportion in the retail trade in today's society. A variety of research on e-commerce is in the ascendant. Recommendation system is one of the important research directions, which has great practical significance for the development of e-commerce. At present, almost all large e-commerce sites, such as Amazon Bay, Dangdang, Excellence, JingDong Mall and so on. At present, the research of E-commerce system is mainly focused on various recommendation technology. There is not much research on the visualization of E-commerce website recommendation. Firstly, this paper studies the status quo of e-commerce website recommendation technology, analyzes and compares the current recommendation technology, and puts forward a recommendation mechanism based on rough set. The recommendation mechanism and algorithm are studied and designed, and the link algorithm between user behavior and purchase intention is obtained by reducing the data of user browsing behavior record through rough set theory. Secondly, this paper analyzes the current research situation. This paper analyzes and tests the current mainstream e-commerce websites based on usability testing technology, and concludes the shortcomings of the current website recommendation visualization system, and designs a new E-commerce website recommendation system. At the same time, the high fidelity prototype of the system is made and the usability evaluation test is carried out. The newly designed recommendation system is verified. In this paper, a large number of usability tests have been done, which provide many suggestions and methods from users for the design of the system. Finally, the verification also shows that the new system has better usability. However, there is not much research on the verification of the algorithm in this paper, and it is also the work that needs to be improved in the future.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2010
【分类号】：J524

【参考文献】