协同过滤技术在高校选课推荐系统中的应用研究

发布时间：2018-12-19 20:03

【摘要】：目前信息化的高校选课系统在高校选课中的应用非常普遍,几乎所有高校都采用了教务网络管理信息系统来为学生提供各种服务。学生使用选课系统在网上进行选课时,可以根据系统中提供的各种选修课,选择自己感兴趣的课程,修满学分完成学业。尽管目前绝大多数的高校选课系统可以实现为学生提供基本的选课功能,但是选课系统的智能化程度不够,不能为学生选课提供推荐功能。学生在选课的时候都存在很大的盲目性,对自己所学专业了解不够,没有方向性。很多同学通常都是为了修满学分,选课的目的性不明确,所选课程是否对自己的专业发展有帮助也不会去考虑。导致所选课程对他们的整个学业规划帮助不大,所以高校教务系统迫切需要有更加智能化的选课系统,系统可以通过智能推荐来解决学生选课时存在的盲目性的问题,方便学生选课,为学生提供更便利的服务。本文主要研究将协同过滤技术融合到高校选课系统中,通过课程之间的相关性,学生之间的相关性,根据学生对所学课程的评分,构建评分矩阵。根据协同过滤算法产生预测评分矩阵,产生推荐列表,实现智能化的为学生推荐他可能喜爱的课程。学生通过系统提供的推荐课程进行选课,所推荐课程往往是学生更感兴趣的,有一定的专业导向性,目的性更明确。学生通过学习选修课程对学生的整个学习体系更加有帮助,达到了通过选修课程来帮助学生扩展知识面,完善专业知识体系的目的。实验结果表明,在传统协同过滤推荐技术的基础上,采用改进的基于项目和用户同时加权的用户协同过滤算法实现课程推荐,推荐精度更高。本文所实现的改进的协同过滤技术在高校选课推荐系统中的应用很好的为学生智能推荐选修课程,为学生所推荐的课程具有很好的合理性和准确性,实现了更加智能化的为学生选课,具有很大的实用性和现实意义。
[Abstract]:At present, the application of the information-based course selection system in colleges and universities is very common. Almost all colleges and universities use the educational administration network management information system to provide various services for students. Students can choose the courses they are interested in according to the various elective courses offered by the system, and complete their studies with credit. Although most of the courses selection systems in colleges and universities can provide students with basic course selection functions, the intelligent degree of the course selection system is not enough to provide the recommended function for students to choose courses. Students in the selection of courses are very blind, their own professional understanding is not enough, there is no direction. Many students are usually to complete the credit, the purpose of the course is not clear, whether the course is helpful to their professional development and will not be considered. As a result, the selected courses are of little help to their entire academic planning. Therefore, the educational administration system in colleges and universities urgently needs a more intelligent course selection system. The system can solve the blindness problems existing in the course selection of students through intelligent recommendation. It is convenient for students to choose classes and provide more convenient service for students. This paper mainly studies the integration of collaborative filtering technology into the course selection system of colleges and universities. Through the correlation between courses and the correlation between students, the scoring matrix is constructed according to the students' scores on the courses they have learned. According to the collaborative filtering algorithm, the prediction score matrix is generated, the recommendation list is generated, and the course that he may like to be recommended to the student is intelligently recommended. Students choose courses through the recommended courses provided by the system. The recommended courses are often more interesting, professional oriented and more purposeful. Through the study of elective courses, students are more helpful to the whole learning system of students, and achieve the goal of helping students expand their knowledge and improve their professional knowledge system through elective courses. The experimental results show that, based on the traditional collaborative filtering recommendation technology, the improved user collaborative filtering algorithm based on item and user weights is used to realize the course recommendation, and the recommendation accuracy is higher. The application of the improved collaborative filtering technology in the course selection recommendation system of colleges and universities is very good for the students to recommend intelligent elective courses, the recommended courses for students have good rationality and accuracy. It realizes more intelligent course selection for students, which has great practicability and practical significance.
【学位授予单位】：华侨大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.3

【参考文献】