论文部分内容阅读
推荐系统的目的是对自动生成的客户提供有针对性的建议。目前,推荐系统虽然得到了很好的研究,但其仍然处于一个发展阶段,并且面临着两大挑战:1)商业应用中的项目和配置文件常常呈现复杂的树状结构,不能由正常项目相似措施来处理2)在线用户的偏好往往是模糊的,不能由现有的推荐方法来处理。为了同时处理这两个挑战,本文提出了一个建立模糊树状结构用户偏好模型的方法,其中模糊集技术被用于表达用户偏好。
The purpose of the recommendation system is to provide targeted advice to automatically generated clients. Currently, the recommendation system, while well studied, is still at a stage of development and faces two major challenges: 1) Projects and profiles in commercial applications often exhibit complex tree structures that can not be similar to normal projects Measures to deal with 2) The preferences of online users are often vague and can not be handled by existing recommendations. In order to deal with both of these challenges simultaneously, this paper presents a method of building a fuzzy tree structure user preference model, in which fuzzy set techniques are used to express user preferences.