Abstract: To find some hidden association rules mainly through historical data, but for different users, different rules are more effective. Thus to find out more valuable rules arouses extensive attention. This paper presents an improved model based on user interest association rules algorithm classification from the perspective of the user to find different rules for different users. The algorithm has been applied in the current teaching evaluation in the universities to verify the validity of the method.

Key words: user recommendation model, FP-TREE, teaching evaluation

摘要： 关联规则主要通过历史数据来找出某些隐性的规律，但是针对不同用户，不同的规则更为有效，因此如何找出比较有价值的规则引起了人们的广泛关注。提出一种基于用户兴趣模型的改进关联规则算法，该算法从用户分类的角度找到适合不同用户的不同规则。通过在学校教学评估中的应用，验证了算法的有效性。

关键词: 用户推荐模型, FP-TREE, 高校教评