Abstract: This paper proposes a user profile representation based on vector space model together with its dynamic learning algorithm, and studies feature selection in user modeling. A new feature selection method combining term frequency and TFIDF according to part-of-speech tagging is proposed. The experiment indicates that the dynamic learning algorithm can catch and record user’s latest interest in time, thus the user required information can be truly recommended. The experiment also shows that the effect of combining method based on part-of-speech tagging is better than that of using TF or TFIDF separately.

Key words: personalized recommending, user profile, feature selection, dynamic updating

摘要： 提出了一种基于向量空间模型的用户模型表示及其动态学习算法，研究了用户建模中的特征选择，提出了一种根据词性标注信息将词频法和TFIDF方法相结合的特征选择方法。实验结果表明这种动态学习算法能实时捕捉并记录用户最新的兴趣需求，从而准确地推荐出符合用户兴趣的信息，同时这种基于词性标注的组合特征选择方法的效果好于单独使用词频法或TFIDF方法。

关键词: 个性化推荐, 用户模型, 特征选择, 动态更新

CLC Number: 

• TP18