基于Web挖掘的个性化教学推荐系统

针对学生网络学习环境设计了一种新颖的个性化教学推荐系统。该系统通过测试学生的学习风格和挖掘Web浏览日志,构造了不同学生学习风格和Web使用习惯的模型。首先利用Item-Based Top-N推荐算法对数据稀疏的学习风格测量数据进行处理,实现对学生学习风格的诊断;然后,采用AprioriAll算法挖掘Web浏览日志中序列频繁集,分析出学生Web使用的常见习惯和兴趣;最后,依据不同的学习风格和Web使用习惯实现学习内容的个性化推荐。模拟实验表明,该推荐系统的设计是可行并有效的,能够很好地符合用户的真实需求。

Personalized Teaching Recommendation System Based on Web Mining

Aiming at the network learning environment for students,we design a new personalized teaching recommendation system.Through testing the learning style of students and mining their Web browsing logs,the system constructs the models with different learning styles and Web use habits.Firstly the system processes the sparse learning style testing data with Item-Based Top-N recommendation algorithm to achieve the diagnosis of students＇ learning style,then it analyzes the habits and interests of students＇ Web use through mining the frequent set of sequences in the Web browsing log with AprioriAll algorithm,finally it realizes the personalized recommendation to the learning content based on different learning styles and Web use habits.The simulated experiment shows that the recommendation model is feasible and effective,and it can well satisfy the real requirements of the users.