基于OJ数据的习题个性化推荐系统-毕业设计论文.doc

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，无法准确地寻找到满足自身学习需求的学习资源，最终导致学习兴趣下降，系统教辅作用无法得到充分地发挥。 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 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，本文提出了推荐算法，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。 Abstract Entered 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-Learning 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 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, there 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 low. For the above phenomenon, in order to better achieve the effect of personalized recommendation system, under the guidance of the theory, we propose a memory-based filtering and rule-based filtering recommendation algorithm, and ultimately the system for student exercis