《基于贝叶斯网络的高校选修课推荐系统的设计与实现(初稿) .doc》由会员分享，可在线阅读，更多相关《基于贝叶斯网络的高校选修课推荐系统的设计与实现(初稿) .doc（80页珍藏版）》请在金锄头文库上搜索。
1、 硕 士 学 位 论 文（专 业 学 位）基于贝叶斯网络的高校选课推荐系统的设计与实现院 系：软件学院专 业： 软件工程姓 名：荣杨指 导 教 师： 王欢完 成 日 期： 2015年09月19日基于贝叶斯网络的高校选课推荐系统的设计与实现 摘要摘 要在高等院校中，教书育人是第一位的任务。所以，教务管理系统在整个高校管理中，扮演了最重要的角色。随着信息技术的发展，教务管理工作信息化的程度越来越普及，要求也越来越高。并且，在历年来的教务管理中，积累了大量的相关教学和管理数据，因此，如何能够结合具体需求，充分开发和挖掘这些数据，并结合高校校务管理方和使用者的具体需求，开发功能更强大，更加人性化的教务
2、系统，从而服务好学生，为教务管理方提供科学支持，就变成一个重要的研究方向。本文以高校教务选课系统为研究课题，结合贝叶斯网络推荐技术，进行了相关研究。为了更好地服务学生群体，为他们的选课提供最佳的咨询和推荐服务，本文利用了个性化推荐技术。其思路主要是采用了贝叶斯网络的推荐技术，本文的主要研究内容如下所示：首先在第一章研究绪论及第二章的相关理论研究部分，综述了研究背景，主要是贝叶斯网络相关研究。介绍了个性化推荐技术以及贝叶斯网络方面的理论基础，并且综述了目前比较流行的推荐引擎的具体算法，以及相关的模型。对这些模型的优点和缺点逐一进行了总结，为后文的研究，奠定了理论基础。然后经过充分调研，了解了某高
3、校在教务管理系统方面的具体需求，重点是该校的选课系统的要求。包括教师、教务管理工作者、学生的具体需求。然后结合三者的需求，设计了高校选课系统的实施方案。接着，结合理论研究部分，设计了基于贝叶斯网络的选课模型以及推荐算法。该算法主要是基于Pearl概率推理思想，在获取学生既往的选课历史后，使用相关的推理与匹配技术，将课程数据库中的科目加以挑选，利用科目的相似度作为度量值，设置为推荐模型的先验概率值。此后，在此基础上，对整个教务系统进行了实现，包括代码实现，测试以及上限和培训。并对整个系统进行了测试。测试结果表明，测试表明，系统能够很好的应用数据分析功能，有效地组织和管理海量的选课数据，在基于学生
4、选课历史的基础上，提供个性化的推荐方案，从而提高了学生选课的效率，减少了盲目性，并且能够为高校教务管理工作者提供更好的数据分析和决策服务。系统充分满足了教务工作者、学生以及教师三个群体的具体需求，功能良好。把基于贝叶斯网络的推荐模型引入到教务选课系统之中，是本文的大胆尝试，目的在于希望教学管理者能够更好地利用信息系统服务学生，实施教学管理的同时，更好的提高学校教学质量和管理效率。关键词：选课系统 贝叶斯网络 推荐模型 个性化推荐 贝叶斯算法基于贝叶斯网络的高校选课推荐系统的设计与实现 AbstractABSTRACTWith the development of information tec
5、hnology and the development of educational informatization, educational management system in Colleges and universities in the daily management of the plays more and more important role and a large number of relevant teaching and management data in educational management over the years accumulated a,
6、 therefore, how to combine the specific needs, fully developed and the data mining, and combined with the specific needs of university affairs management and users, development of more powerful, more human nature of academic system, so as to serve the students, in order to provide scientific support
7、 for educational management, becomes an important research direction. In this paper, the university educational administration system for the study, combined with the Bayesian network recommended technology, the related research.In order to better serve the student population, to provide the best ad
8、vice and service for their courses, this paper uses the personalized recommendation technology. The main research contents of this paper are as follows: firstly, the relevant theoretical research of the first chapter is the introduction and the two chapter, the research background, the main research
9、 background, the research of Bayesian network. This paper introduces the theory of personalized recommendation technology and the theory of Bayesian network, and summarizes the specific algorithms of the popular recommendation engine, and the relevant model. The advantages and disadvantages of these
10、 models are summarized, which lay a theoretical foundation for the study of the paper. Then through the full investigation and research, the specific needs of the educational management system in a university is understood, and the emphasis is on the requirements of the system. The specific needs of
11、 the teachers, the educational administration management staff and the students. And then combined with the needs of the three, the design of the university elective system implementation plan. Then, based on the theory research, the selection of the model and the recommendation algorithm based on B
12、ayesian network is designed. This algorithm is based on the Pearl probability reasoning, in order to obtain the students past history, use the relevant reasoning and matching technology, the subjects in the database to be selected, the use of the similarity of subjects as a measure, set the recommen
13、ded model of a priori probability. Thereafter, on this basis, the entire educational system was implemented, including code implementation, testing, and training. And the whole system is tested. The test results show that the system can be well applied to data analysis function, effectively organize
14、 and manage the massive data, based on the history of students selecting courses, to provide personalized recommendation, which can improve the efficiency of the students choosing courses, reduce the blindness, and can provide better data analysis and decision-making service for the educational admi
15、nistration management. The system fully meets the specific needs of the educational workers, students and teachers of the three groups, the function is good.The recommendation model based on Bayesian network is introduced into the educational administration system, which is the purpose of this paper is to hope that the students can better use the information system to serve the students, implement the teaching management and improve the teaching quality and management efficiency.Key words: course selection system, Bias network, recommendation model, personalized recommend
