基于偏好的教育推荐系统的业务逻辑设计

发布时间：2018-05-29 18:35

本文选题：业务逻辑 + 推荐系统　；参考：《中南大学》2013年硕士论文

【摘要】：互联网上数据的爆炸式增长,直接催生了信息过载。数据量急速增长的一个重要原因是资源访问的无限制性。在资源搜索时信息交换的巨大性是一个需要考虑的问题。用户花费大量时间和精力来查找可靠的并与其需求相关的信息,因为搜索引擎是通过资源的点击率、重要性等对检索结果进行排序,而没有考虑到用户兴趣。研究表明搜索引擎的检索效率仍有待提高。当前已经有一些信息检索与过滤的方法被研究人员提出来,但其要么需要用户大量的反馈信息,要么需要用户花费大量时间来对系统进行学习。时间对用户来说是极其宝贵的,尤其对学生来说,现有的学习系统几乎没有考虑到学生的偏好。系统提供给用户的资源不仅应该考虑用户的偏好信息,同时也应该将其他用户对该资源的评价考虑在内。一个比较可行的方案是融合用户的偏好与相似用户对资源的评价信息。业务逻辑设计是系统的核心部件。它负责端到端的数据交互以及为用户接口和数据库之间架设桥梁。业务逻辑中定义的规则集合管理整个系统的数据流程,系统的可靠性与高效性与业务逻辑设计息息相关。本文中我们提出了一个教育推荐系统业务逻辑设计模型,其中考虑到用户的偏好、资源的权重以及资源当时的热度。我们提出了一个查找相似用户的算法,在推荐系统中,相似性用户的查找极其重要,查找到相似性用户后还需要考虑的一个方面是用户的权重。系统根据用户的信息推荐一个资源序列给用户,资源的排名依据用户的相邻用户信息融合权重以及资源的热度。因此本文主要解决的问题是提出一个教育推荐系统业务逻辑设计模型,其根据用户的偏好、资源的权重、资源的热度以及用户的相邻用户信息向用户推荐教育资源。图4副,表3个,参考文献60篇。
[Abstract]:The explosive growth of data on the Internet directly spawned information overload. An important reason for the rapid growth of data is the unrestricted access to resources. The huge nature of information exchange in resource search is a problem to be considered. Users spend a lot of time and effort to find reliable and relevant information because search engines sort the retrieval results through the click rate and importance of resources without considering the interests of users. The research shows that the search efficiency of search engine still needs to be improved. At present, some methods of information retrieval and filtering have been proposed by researchers, but they either need a lot of feedback from users, or they need a lot of time to study the system. Time is extremely valuable to users, especially for students, the existing learning system hardly takes students' preferences into account. The resources provided by the system should not only consider the user's preference information, but also take into account the evaluation of the resource by other users. A more feasible scheme is to combine user preferences with resource evaluation information of similar users. Business logic design is the core part of the system. It is responsible for end-to-end data interaction and building bridges between user interfaces and databases. The set of rules defined in business logic manages the data flow of the whole system. The reliability and efficiency of the system are closely related to the design of business logic. In this paper, we propose a business logic design model for educational recommendation system, which takes into account the user's preference, the weight of the resource and the heat of the resource at that time. We propose an algorithm to find similar users. In the recommendation system, the search of similar users is extremely important, and one of the aspects to be considered after finding similar users is the weight of users. The system recommends a resource sequence to the user according to the information of the user. The ranking of the resource is based on the information fusion weight of the user and the heat of the resource. Therefore, the main problem of this paper is to propose a business logic design model of educational recommendation system, which recommends educational resources to users according to the user's preference, the weight of resources, the heat of resources and the information of users' adjacent users. Fig4 pairs, 3 tables, 60 references.
【学位授予单位】：中南大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP391.3

【共引文献】