个性化推荐系统毕业设计论文

《个性化推荐系统毕业设计论文》由会员分享，可在线阅读，更多相关《个性化推荐系统毕业设计论文（66页珍藏版）》请在毕设资料网上搜索。

1、 摘要摘要 进入大数据的时代，信息的产生、加工、传播变得越来越容易，计算 机已经深入到人们现代生活的方方面面，在教育领域，智能教学系统作为 一种辅助教学的手段逐渐成为 E-Learning 领域众多学者的研究重点，但多 数智能教学系统缺乏有效的学习资源推荐机制，使得用户无法准确地寻找 到满足自身学习需求的学习资源，最终导致用户学习兴趣下降，系统教辅 作用无法得到充分地发挥。 个性化推荐是缓建信息过载的方法之一，为了解决在海量题库中为学 生提供个性化资源推荐的问题，对于传统的基于用户的协同过滤和基于内 容的协同过滤，存在很多弊端，难以满足用户的需求。对于基于用户的协 同过滤推荐的原则，如果一个用

2、户没有相同喜好的朋友，那该算法的推荐 效果总是不令人满意；对于基于内容的协同过滤推荐的原则，就是假设指 用户会喜欢和他以前喜欢的东西是比较相似的，如果不是相似的，用该算 法做出准确的推荐的可能性非常低。 针对以上现象，为了能够更好的达到个性化推荐系统的效果，在相关 理论的指导下，本文提出了基于记忆的过滤和基于规则的过滤两种推荐算 法，最终实现习题系统对于学生用户的个性化推荐，以避免传统推荐系统 因忽略资源本身蕴涵信息而产生的无关推荐，为智能教学系统中的应用提 供一种新的发展思路。 关键词关键词：个性化推荐，基于用户的协同过滤，基于内容的协同过滤，基于 规则的过滤 Abstract Entere

3、d the time of Big Data, the generation, processing, dissemination of information getting easier. Computer has penetrated into every aspect of modern life , In the field of education , intelligent tutoring system as a means of secondary education is becoming the focus of many scholars study E-Learnin

4、g in the field , but most intelligent tutoring system, the lack of effective mechanisms recommended learning resources , making the users can not exactly looking to meet their learning needs of learning resources , resulting in decreased interest users to learn the system and supplementary role can

5、not be fully realized. Personalized recommendation is postponed one of the methods of information overload, in order to solve the massive exam to provide personalized resources recommended questions for the students, the traditional collaborative filtering and user-based content-based filtering, the

6、re are many drawbacks, it is difficult to meet the needs of users. For the principle of user-based collaborative filtering recommendation, if a user does not have the same taste in friends, it is recommended that the algorithm is undesirable of man; for content-based collaborative filtering recommendation principle, means users will like his former favorite things relatively similar, if the self-similarity is small, the possibility of using this method to make accurate recommendation is very lo